Multiple module computer system and method including differential signal channel comprising unidirectional serial bit channels to transmit encoded peripheral component interconnect bus transaction data

CROSS REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. application Ser. No. 12/077,503 filed Mar. 18, 2008 now U.S. Pat. No. 7,676,624, which is a continuation of U.S. nonprovisional application Ser. No. 11/166,656, filed Jun. 24, 2005 (Now U.S. Pat. No. 7,376,779), which is a continuation of U.S. nonprovisional application Ser. No. 11/097,694, filed Mar. 31, 2005 (Now U.S. Pat. No. 7,363,415), which is a continuation of U.S. nonprovisional application Ser. No. 10/772,214, filed Feb. 3, 2004 (Now U.S. Pat. No. 7,099,981), which is a continuation of U.S. nonprovisional application Ser. No. 09/569,758, filed May 12, 2000 (Now U.S. Pat. No. 6,718,415), and which claimed priority to U.S. Provisional Application No. 60/134,122 filed May 14, 1999. These applications are hereby incorporated by reference in their entirety.

BACKGROUND OF THE INVENTION

The present invention relates to computing devices. More particularly, the present invention provides a system including a plurality of computer modules that can independently operate to provide backup capability, dual processing, and the like. Merely by way of example, the present invention is applied to a modular computing environment for desk top computers, but it will be recognized that the invention has a much wider range of applicability. It can be applied to a server as well as other portable or modular computing applications.

Many desktop or personal computers, which are commonly termed PCs, have been around and used for over ten years. The PCs often come with state-of-art microprocessors such as the Intel Pentium™ microprocessor chips. They also include a hard or fixed disk drive such as memory in the giga-bit range. Additionally, the PCs often include a random access memory integrated circuit device such as a dynamic random access memory device, which is commonly termed DRAM. The DRAM devices now provide up to millions of memory cells (i.e., mega-bit) on a single slice of silicon. PCs also include a high resolution display such as cathode ray tubes or CRTs. In most cases, the CRTs are at least 15 inches or 17 inches or 20 inches in diameter. High resolution flat panel displays are also used with PCs.

Many external or peripheral devices can be used with the PCs. Among others, these peripheral devices include mass storage devices such as a Zip™ Drive product sold by Iomega Corporation of Utah. Other storage devices include external hard drives, tape drives, and others. Additional devices include communication devices such as a modem, which can be used to link the PC to a wide area network of computers such as the Internet. Furthermore, the PC can include output devices such as a printer and other output means. Moreover, the PC can include special audio output devices such as speakers the like.

PCs also have easy to use keyboards, mouse input devices, and the like. The keyboard is generally configured similar to a typewriter format. The keyboard also has the length and width for easily inputting information by way of keys to the computer. The mouse also has a sufficient size and shape to easily move a cursor on the display from one location to another location.

Other types of computing devices include portable computing devices such as “laptop” computers and the like. Although somewhat successful, laptop computers have many limitations. These computing devices have poor display technology. In fact, these devices often have a smaller flat panel display that has poor viewing characteristics. Additionally, these devices also have poor input devices such as smaller keyboards and the like. Furthermore. these devices have limited common platforms to transfer information to and from these devices and other devices such as PCs.

Up to now, there has been little common ground between these platforms including the PCs and laptops in terms of upgrading, ease-of-use, cost, performance, and the like. Many differences between these platforms, probably somewhat intentional, has benefited computer manufacturers at the cost of consumers. A drawback to having two separate computers is that the user must often purchase both the desktop and laptop to have “total” computing power, where the desktop serves as a “regular” computer and the laptop serves as a “portable” computer. Purchasing both computers is often costly and runs “thousands” of dollars. The user also wastes a significant amount of time transferring software and data between the two types of computers. For example, the user must often couple the portable computer to a local area network (i.e., LAN), to a serial port with a modem and then manually transfer over files and data between the desktop and the portable computer. Alternatively, the user often must use floppy disks to “zip” up files and programs that exceed the storage capacity of conventional floppy disks, and transfer the floppy disk data manually.

Another drawback with the current model of separate portable and desktop computer is that the user has to spend money to buy components and peripherals the are duplicated in at least one of these computers. For example, both the desktop and portable computers typically include hard disk drives, floppy drives, CD-ROMs, computer memory, host processors, graphics accelerators, and the like. Because program software and supporting programs generally must be installed upon both hard drives in order for the user to operate programs on the road and in the office, hard disk space is often wasted.

One approach to reduce some of these drawbacks has been the use of a docking station with a portable computer. Here, the user has the portable computer for “on the road” use and a docking station that houses the portable computer for office use.

Similar to separate desktop and portable computers, there is no commonality between two desktop computers. To date, most personal computers are constructed with a single motherboard that provides connection for CPU and other components in the computer. Dual CPU systems have been available through Intel's slot 1 architecture. For example, two Pentium II cartridges can be plugged into two “slot 1” card slots on a motherboard to form a Dual-processor system. The two CPU's share a common host bus that connects to the rest of the system, e.g. main memory, hard disk drive, graphics subsystem, and others. Dual CPU systems have the advantage of increased CPU performance for the whole system. Adding a CPU cartridge requires no change in operating systems and application software. However, dual CPU systems may suffer limited performance improvement if memory or disk drive bandwidth becomes the limiting factor. Also, dual CPU systems have to time-share the processing unit in running multiple applications. CPU performance improvement efficiency also depends on software coding structure. Dual CPU systems provide no hardware redundancy to help fault tolerance. In running multiple applications, memory and disk drive data throughput will become the limiting factor in improving performance with multi-processor systems.

The present invention generally relates to computer interfaces. More specifically, the present invention relates to an interface channel that interfaces two computer interface buses that operate under protocols that are different from that used by the interface channel.

Interfaces coupling two independent computer buses are well known in the art. A block diagram of a computer system utilizing such a prior art interface is shown in <figref idref="DRAWINGS">FIG. 5</figref>. In <figref idref="DRAWINGS">FIG. 5</figref>, a primary peripheral component interconnect (PCI) bus <b>505</b> of a notebook PC <b>500</b> is coupled to a secondary PCI bus <b>555</b> in a docking system <b>550</b> (also referred to as docking station <b>550</b>) through high pin count connectors <b>501</b> and <b>502</b>, which are normally mating connectors. The high pin count connectors <b>501</b> and <b>502</b> contain a sufficiently large number of pins so as to carry PCI bus signals between the two PCI buses without any translation. The main purpose for interfacing the two independent PCI buses is to allow transactions to occur between a master on one PCI bus and a target on the other PCI bus. The interface between these two independent PCI buses additionally includes an optional PCI to PCI bridge <b>560</b>, located in the docking station <b>550</b>, to expand the add on capability in docking station <b>550</b>. The bridge <b>560</b> creates a new bus number for devices behind the bridge <b>560</b> so that they are not on the same bus number as other devices in the system thus increasing the add on capability in the docking station <b>550</b>.

An interface such as that shown in <figref idref="DRAWINGS">FIG. 5</figref> provides an adequate interface between the primary and secondary PCI buses. However, the interface is limited in a number of ways. The interface transfers signals between the primary and secondary PCI buses using the protocols of a PCI bus. Consequently, the interface is subject to the limitations under which PCI buses operate. One such limitation is the fact that PCI buses are not cable friendly. The cable friendliness of the interface was not a major concern in the prior art. However, in the context of the computer system of the present invention, which is described in the present inventor's (William W. Y. Chu's) application for “Personal Computer Peripheral Console With Attached. Computer Module” filed concurrently with the present application on Sep. 8, 1998 and incorporated herein by reference, a cable friendly interface is desired for interfacing an attached computer module (ACM) and a peripheral console of the present invention. Furthermore, as a result of operating by PCI protocols, the prior art interface includes a very large number of signal channels with a corresponding large number of conductive lines (and a similarly large number of pins in the connectors of the interface) that are commensurate in number with the number of signal lines in the PCI buses which it interfaces. One disadvantage of an interface having a relatively large number of conductive lines and pins is that it costs more than one that uses a fewer number of conductive lines and pins. Additionally, an interface having a large number of conductive lines is bulkier and more cumbersome to handle. Finally, a relatively large number of signal channels in the interface renders the option of using differential voltage signals less viable because a differential voltage signal method would require duplicating a large number of signal lines. It is desirable to use a low voltage differential signal (LVDS) channel in the computer system of the present invention because an LVDS channel is more cable friendly, faster, consumes less power, and generates less noise, including electromagnetic interferences (EMI), than a PCI channel. The term LVDS is herein used to generically refer to low voltage differential signals and is not intended to be limited to any particular type of LVDS technology.

Thus, what is needed are computer systems that can have multiple computer modules. Each computer module has dedicated memory and disk drive, and can operate independently.

BRIEF SUMMARY OF THE INVENTION

According to the present invention, a technique including a method and device for multi-module computing is provided. In an exemplary embodiment, the present invention provides a system including a plurality of computer modules that can independently operate to provide backup capability, dual processing, and the like.

In a specific embodiment, the present invention provides a computer system for multi-processing purposes. The computer system has a console comprising a first coupling site and a second coupling site, e.g., computer module bay. Each coupling site comprises a connector. The console is an enclosure that is capable of housing each coupling site. The system also has a plurality of computer modules, where each of the computer modules is coupled to one of the connectors. Each of the computer modules has a processing unit, a main memory coupled to the processing unit, a graphics controller coupled to the processing unit, and a mass storage device coupled to the processing unit. Each of the computer modules is substantially similar in design to each other to provide independent processing of each of the computer modules in the computer system.

In an alternative specific embodiment, the present invention provides a multi-processing computer system. The system has a console comprising a first coupling site and a second coupling site. Each coupling site comprises a connector. The console is an enclosure that is capable of housing each coupling site. The system also has a plurality of computer modules, where each of the computer modules is coupled to one of the connectors. Each of the computer modules has a processing unit, a main memory coupled to the processing unit, a graphics controller coupled to the processing unit, a mass storage device coupled to the processing unit, and a video output coupled to the processing unit. Each of the computer modules is substantially similar in design to each other to provide independent processing of each of the computer modules in the computer system. A video switch circuit is coupled to each of the computer modules through the video output. The video switch is configured to switch a video signal from any one of the computer modules to a display.

Numerous benefits are achieved using the present invention over previously existing techniques. In one embodiment, the invention provides improved processing and maintenance features. The invention can also provide increased CPU performance for the whole system. The invention also can be implemented without changes in operating system and application software. The present invention is also implemented using conventional technologies that can be provided in the present computer system in an easy and efficient manner.

In another embodiment, the invention provides at least two users to share the same modular desktop system. Each user operates on a different computer module. The other peripheral devices, i.e. CDROM, printer, DSL connection, etc. can be shared. This provides lower system cost, less desktop space and more efficiency. Depending upon the embodiment, one or more of these benefits can be available. These and other advantages or benefits are described throughout the present specification and are described more particularly below.

In still further embodiments, the present invention provides methods of using multiple computer modules.

The present invention encompasses an apparatus for bridging a first computer interface bus and a second computer interface bus. where each of the first and second computer interface buses have a number of parallel multiplexed address/data bus lines and operate at a clock speed in a predetermined clock speed range having a minimum clock speed and a maximum clock speed. The apparatus comprises an interface channel having a clock line and a plurality of bit lines for transmitting bits; a first interface controller coupled to the first computer interface bus and to the interface channel to encode first control signals from the first computer interface bus into first control bits to be transmitted on the interface channel and to decode second control bits received from the interface channel into second control signals to be transmitted to the first computer interface bus; and a second interface controller coupled to the interface channel and the second computer interface bus to decode the first control bits from the interface channel into third control signals to be transmitted on the second computer interface bus and to encode fourth control signals from the second computer interface bus into the second control bits to be transmitted on the interface channel.

In one embodiment, the first and second interface controllers comprise a host interface controller (HIC) and a peripheral interface controller (PIC), respectively, the first and second computer interface buses comprise a primary PCI and a secondary PCI bus, respectively, and the interface channel comprises an LVDS channel.

The present invention overcomes the aforementioned disadvantages of the prior art by interfacing two PCI or PCI-like buses using a non-PCI or non-PCI-like channel. In the present invention, PCI control signals are encoded into control bits and the control bits, rather than the control signals that they represent, are transmitted on the interface channel. At the receiving end, the control bits representing control signals are decoded back into PCI control signals prior to being transmitted to the intended PCI bus.

The fact that control bits rather than control signals are transmitted on the interface channel allows using a smaller number of signal channels and a correspondingly small number of conductive lines in the interface channel than would otherwise be possible. This is because the control bits can be more easily multiplexed at one end of the interface channel and recovered at the other end than control signals. This relatively small number of signal channels used in the interface channel allows using LVDS channels for the interface. As mentioned above, an LVDS channel is more cable friendly, faster, consumes less power, and generates less noise than a PCI bus channel, which is used in the prior art to interface two PCI buses. Therefore, the present invention advantageously uses an LVDS channel for the hereto unused purpose of interfacing PCI or PCI-like buses. The relatively smaller number of signal channels in the interface also allows using connectors having smaller pins counts. As mentioned above an interface having a smaller number of signal channels and, therefore, a smaller number of conductive lines is less bulky and less expensive than one having a larger number of signal channels. Similarly, connectors having a smaller number of pins are also less expensive and less bulky than connectors having a larger number of pins.

In one embodiment, the present invention encompasses an apparatus for bridging a first computer interface bus and a second computer interface bus, in a microprocessor based computer system where each of the first and second computer interface buses have a number of parallel multiplexed address/data bus lines and operate at a clock speed in a predetermined clock speed range having a minimum clock speed and a maximum clock speed. The apparatus comprises an interface channel having a clock channel and a plurality of bit channels for transmitting bits; a first interface controller coupled to the first computer interface bus and to the interface channel to encode first control signals from the first computer interface bus into first control bits to be transmitted on the interface channel and to decode second control bits received from the interface channel into second control signals to be transmitted to the first computer interface bus; and a second interface controller coupled to the interface channel and the second computer interface bus to decode the first control bits from the interface channel into third control signals to be transmitted on the second computer interface bus and to encode fourth control signals from the second computer interface bus into the second control bits to be transmitted on the interface channel.

In one embodiment, the first and second interface controllers comprise a host interface controller (HIC) and a peripheral interface controller (PIC), respectively, the first and second computer interface buses comprise a primary PCI and a secondary PCI bus, respectively, and the interface channel comprises an LVDS channel.

In a preferred embodiment, the interface channel has a plurality of serial bit channels numbering fewer than the number of parallel bus lines in each of the PCI buses and operates at a clock speed higher than the clock speed at which any of the bus lines operates. More specifically, the interface channel includes two sets of unidirectional serial bit channels which transmit data in opposite directions such that one set of bit channels transmits serial bits from the HIC to the PIC while the other set transmits serial bits from the PIC to the HIC. For each cycle of the PCI clock, each bit channel of the interface channel transmits a packet of serial bits.

The HIC and PIC each include a bus controller to interface with the first and second computer interface buses, respectively, and to manage transactions that occur therewith. The HIC and PIC also include a translator coupled to the bus controller to encode control signals from the first and second computer interface buses, respectively, into control bits and to decode control bits from the interface channel into control signals. Additionally, the HIC and PIC each include a transmitter and a receiver coupled to the translator. The transmitter converts parallel bits into serial bits and transmits the serial bits to the interface channel. The receiver receives serial bits from the interface channel and converts them into parallel bits.

According to the present invention, a technique including a method and device for securing a computer module using a password in a computer system is provided. In an exemplary embodiment, the present invention provides a security system for an attached computer module (“ACM”). In an embodiment, the ACM inserts into a Computer Module Bay (CMB) within a peripheral console to form a functional computer.

In a specific embodiment, the present invention provides a computer module. The computer module has an enclosure that is insertable into a console. The module also has a central processing unit (i.e., integrated circuit chip) in the enclosure. The module has a hard disk drive in the enclosure, where the hard disk drive is coupled to the central processing unit. The module further has a programmable memory device in the enclosure, where the programmable memory device can be configurable to store a password for preventing a possibility of unauthorized use of the hard disk drive and/or other module elements. The stored password can be any suitable key strokes that a user can change from time to time. In a further embodiment, the present invention provides a permanent password or user identification code stored in flash memory, which also can be in the processing unit, or other integrated circuit element. The permanent password or user identification code is designed to provide a permanent “finger print” on the attached computer module.

In a specific embodiment, the present invention provides a variety of methods. In one embodiment, the present invention provides a method for operating a computer system such as a modular computer system and others. The method includes inserting an attached computer module (“ACM”) into a bay of a modular computer system. The ACM has a microprocessor unit (e.g. microcontroller, microprocessor) coupled to a mass memory storage device (e.g., hard disk). The method also includes applying power to the computer system and the ACM to execute a security program, which is stored in the mass memory storage device. The method also includes prompting for a user password from a user on a display (e.g., flat panel, CRT). In a further embodiment, the present method includes a step of reading a permanent password or user identification code stored in flash memory, or other integrated circuit element. The permanent password or user identification code provides a permanent finger print on the attached computer module. The present invention includes a variety of these methods that can be implemented in computer codes, for example, as well as hardware.

Numerous benefits are achieved using the present invention over previously existing techniques. The present invention provides mechanical and electrical security systems to prevent theft or unauthorized use of the computer system in a specific embodiment. Additionally, the present invention substantially prevents accidental removal of the ACM from the console. In some embodiments, the present invention prevents illegal or unauthorized use during transit. The present invention is also implemented using conventional technologies that can be provided in the present computer system in an easy and efficient manner. Depending upon the embodiment, one or more of these benefits can be available. These and other advantages or benefits are described throughout the present specification and are described more particularly below.

These and other embodiments of the present invention, as well as its advantages and features, are described in more detail in conjunction with the text below and attached Figs.

BRIEF DESCRIPTION OF THE DRAWINGS

<figref idref="DRAWINGS">FIG. 1</figref> is a simplified diagram of a computer system according to an embodiment of the present invention;

<figref idref="DRAWINGS">FIG. 2</figref> is a simplified block diagram of a computer system according to an alternative embodiment of the present invention;

<figref idref="DRAWINGS">FIG. 3</figref> is a simplified block diagram of a compeer system according to a further alternative embodiment of the present invention, and

<figref idref="DRAWINGS">FIG. 4</figref> is a simplified flow diagram of a method according to an embodiment of the present invention.

<figref idref="DRAWINGS">FIG. 5</figref> is a block diagram of a computer system using a prior art interface between a primary and a secondary PCI bus.

<figref idref="DRAWINGS">FIG. 6</figref> is a block diagram of one embodiment of a computer system using the interface of the present invention.

<figref idref="DRAWINGS">FIG. 7</figref> is a partial block diagram of a computer system using the interface of the present invention as a bridge between the north and south bridges of the computer system.

<figref idref="DRAWINGS">FIG. 8</figref> is a partial block diagram of a computer system in which the north and south bridges are integrated with the host and peripheral interface controllers, respectively.

<figref idref="DRAWINGS">FIG. 9</figref> is a block diagram of one embodiment of the host interface controller and the peripheral interface controller of the present invention.

<figref idref="DRAWINGS">FIG. 10</figref> is a detailed block diagram of one embodiment of the host interface controller of the present invention.

<figref idref="DRAWINGS">FIG. 11</figref> is a detailed block diagram of one embodiment of the PIC of the present invention.

<figref idref="DRAWINGS">FIG. 12</figref> is a table showing the symbols, signals, data rate and description of signals in a first embodiment of the XPBus.

<figref idref="DRAWINGS">FIG. 13</figref> is a table showing the information transmitted on the XPBus during two clock cycles of the XPBus in one embodiment of the present invention where 10 data bits transmitted in each clock cycle of the XPBus.

<figref idref="DRAWINGS">FIG. 14</figref> is a table showing information transmitted on the XPBus during four clock cycles of the XPBus in another embodiment of the present invention where 10 data bits are transmitted in each clock cycle of the XPBus.

<figref idref="DRAWINGS">FIG. 15</figref> is a schematic diagram of the signal lines PCK, PD<b>0</b> to PD<b>3</b>, and PCN.

<figref idref="DRAWINGS">FIG. 16</figref> is a table showing the names, types, number of pins dedicated to, and the description of the primary bus PCI signals.

<figref idref="DRAWINGS">FIG. 17</figref> is a block diagram of one embodiment of a computer system employing the present invention.

<figref idref="DRAWINGS">FIG. 18</figref> is a block diagram of an attached computing module (ACM).

<figref idref="DRAWINGS">FIG. 19</figref> illustrates an external view of one embodiment of an ACM.

<figref idref="DRAWINGS">FIG. 19</figref><i>b </i>illustrates one possible embodiment of a computer bay.

<figref idref="DRAWINGS">FIG. 20</figref> illustrates the internal component layout for one embodiment of an ACM.

<figref idref="DRAWINGS">FIG. 21</figref> is a block diagram of a peripheral console (PCON).

<figref idref="DRAWINGS">FIG. 22</figref> is a simplified layout diagram of a security system for a computer system according to an embodiment of the present invention; and

<figref idref="DRAWINGS">FIG. 23</figref> is a simplified block diagram of a security system for a computer module according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

According to the present invention, a technique including a method and device for multi-module computing is provided. In an exemplary embodiment, the present invention provides a system including a plurality of computer modules that can independently operate to provide backup capability, dual processing, and the like.

<figref idref="DRAWINGS">FIG. 1</figref> is a simplified diagram of a computer system <b>100</b> according to an embodiment of the present invention. This diagram is merely an illustration and should not limit the scope of the claims herein. One of ordinary skill in the art would recognize other variations, modifications, and alternatives. The computer system <b>100</b> includes an attached computer module (i.e., ACM) <b>113</b>, a desktop console <b>101</b>, among other elements. The computer system also has another ACM module <b>117</b>. Each ACM module has a respective slot <b>121</b>, <b>119</b>, which mechanically houses and electrically couples each ACM to the computer console. Also shown is a display <b>111</b>, which connects to the console. Additionally, keyboard <b>109</b> and mouse <b>115</b> are also shown. A second display <b>102</b>, keyboard <b>105</b>, and mouse <b>107</b> can be coupled to the console in some optional embodiments to allow more than one user to operate the computer system. The computer system is modular and has a variety of components that are removable. Some of these components (or modules) can be used in different computers, workstations, computerized television sets, and portable or laptop units.

In the present embodiment, each ACM <b>113</b> includes computer components. as will be described below, including a central processing unit (“CPU”). IDE controller, hard disk drive, computer memory, and the like. The computer module bay (i.e., CMB) <b>121</b> is an opening or slot in the desktop console. The CMB houses the ACM and provides communication to and from the ACM. The CMB also provides mechanical protection and support to the ACM. The CMB has a mechanical alignment mechanism for mating a portion of the ACM to the console. The CMB further has thermal heat dissipation sinks, electrical connection mechanisms, and the like. Some details of the ACM can be found in co-pending patent application Ser. Nos. 09/149,882 and 09/149,548 filed Sep. 8, 1998, commonly assigned, and hereby incorporated by reference for all purposes.

In a specific embodiment, the present multiple computer module system has a peripheral console that has two or more computer bays that can receive a removable computer module or ACM. Multiple computer module system can function as a personal computer with only one ACM and the peripheral console. The second and additional ACM can be added later to increase overall system performance and reliability. The ACM operates independently as self-contained computer, communicates with each other through a high-speed serial communication and share most peripheral devices within the peripheral console. Each ACM controls its independent graphics subsystem and drives separate video output signals. A practical implementation is a dual ACM system. In a dual ACM system, two monitors can be used to display the two ACMs' graphics outputs at the same time. For a single monitor, a RGB switch is used to switch between the video outputs of the two ACMs and can be controlled by a command from the user. Similarly, input devices (i.e. keyboard and mouse) are switched between the two computer systems with a command from the user. Command from the user can be in the form of either a dedicated key on the keyboard or a special icon on the screen that the mouse can click on.

In most embodiments, the ACM includes an enclosure such as the one described with the following components, which should not be limiting:

1) A CPU with cache memory;

2) Core logic device or means;

3) Main memory;

4) A single primary Hard Disk Drive (“HDD”) that has a security program;

5) Flash memory with system BIOS and programmable user password;

6) Operating System, application software, data files on primary HDD;

7) An interface device and connectors to peripheral console;

8) A software controllable mechanical lock, lock control means, and other accessories.

The ACM connects to a peripheral console with power supply, a display device, an input device, and other elements. Some details of these elements with the present system are described in more detail below. In a dual ACM system, the primary ACM can connect directly to the peripheral board in the peripheral console. The second ACM can connect either directly or indirectly to the peripheral board. For indirect connection, a receptacle board is added to allow a cable connection to the peripheral board. This is to facilitate the mechanical positioning of the second ACM inside the computer chassis. The receptacle board approach can even be used for the primary ACM if a high bandwidth peripheral bus, e.g. PCI Bus, is not connected from the primary ACM to the peripheral board.

The shared peripheral console has a chassis and a motherboard that connects the following devices:

1) Input means, e.g. keyboard and mouse,

2) Display means, e.g. RGB monitor,

3) Add-on means, e.g. PCI add-on slots,

4) Two Computer Module Bays (CMB) with connectors to two ACMs,

5) A serial communication Hub controller that interfaces to serial communication controller of both ACMs,

6) Shared storage subsystem, e.g. Floppy drive, CDROM drive, DVD drive, or 2nd Hard Drive,

7) Communication device, e.g. modem,

8) Power supply, and others.

The computer bay is an opening in the peripheral console that receives an ACM. CMB provides mechanical protection to ACM, mechanical alignment for connector mating, mechanical locking system to prevent theft and accidental removal, and connectors at the end of the opening for connecting to ACM. The interface bus between ACM and the peripheral console has a video bus, peripheral connections, serial communication connection, control signals and power connection. Video bus includes video output of graphics devices, i.e. analog RGB and control signals for monitor. Power connection supplies the power for ACM.

An implementation of peripheral sharing is the use of Ethernet controllers to bridge the communication between the two ACMs. Some of the peripheral devices residing in the peripheral console are shown in the simplified diagram of <figref idref="DRAWINGS">FIG. 2</figref>. As shown, the diagram is merely an illustration which should not limit the scope of the claims herein. One of ordinary skill in the art would recognize many other variations, alternatives, and modifications. As shown, a primary ACM <b>203</b> is connected to PCI peripheral devices in the peripheral console through the PCI bus <b>225</b> that passes through the connection between primary ACM <b>203</b> and peripheral console <b>201</b>. As shown, ACM has a CPU module <b>207</b> coupled to the PCI bus through a North Bridge <b>211</b>.

The CPU module can use a suitable microprocessing unit, microcontroller, digital signal processor, and the like. In a specific embodiment, the CPU module uses, for example, a 400 MHz Pentium II microprocessor module from Intel Corporation and like microprocessors from AMD Corporation, Cyrix Corporation (now National Semiconductor Corporation), and others. In other aspects, the microprocessor can be one such as the Compaq Computer Corporation Alpha Chip, Apple Computer Corporation PowerPC G3 processor, and the like. Further, higher speed processors are contemplated in other embodiments as technology increases in the future.

In the CPU module, peripheral controller <b>213</b> is coupled to BIOS/flash memory <b>217</b>. Additionally, the peripheral controller is coupled to a clock control logic, a configuration signal, and a peripheral bus. The ACM has the hard drive module <b>215</b>. Among other elements, the ACM includes north bridge <b>215</b>, graphics subsystem <b>223</b> (e.g., graphics accelerator, graphics memory), an IDE controller, and other components. Adjacent to and in parallel alignment with the hard drive module <b>215</b> is the PCI bus. In a specific embodiment, North Bridge unit <b>211</b> often couples to a computer memory <b>209</b>, to the graphics subsystem, and to the peripheral controller via the PCI bus. Graphics subsystem typically couples to a graphics memory, and other elements. IDE controller generally supports and provides timing signals necessary for the IDE bus. In the present embodiment, the IDE controller is embodied as part of a P114XE controller from Intel, for example. Other types of buses than IDE are contemplated, for example EIDE, SCSI, 1394, and the like in alternative embodiments of the present invention.

The hard drive module or mass storage unit <b>215</b> typically includes a computer operating system, application software program files, data files, and the like. In a specific embodiment, the computer operating system may be the Windows98 operating system from Microsoft Corporation of Redmond Wash. Other operating systems, such as WindowsNT, MacOS8, Unix, and the like are also contemplated in alternative embodiments of the present invention. Further, some typical application software programs can include Office98 by Microsoft Corporation, Corel Perfect Suite by Corel, and others. Hard disk module <b>215</b> includes a hard disk drive. The hard disk drive, however, can also be replaced by removable hard disk drives, read/write CD ROMs, flash memory, floppy disk drives, and the like. A small form factor, for example 2.5″, is currently contemplated, however, other form factors, such as PC card, and the like are also contemplated. Mass storage unit <b>240</b> may also support other interfaces than IDE.

Among other features, the computer system includes an ACM with security protection.

The ACM also has a network controller, which can be an Ethernet controller <b>219</b>, which is coupled to the North Bridge through the PCI bus. The North Bridge is coupled to the CPU. The Ethernet controller can be a 10/100 Base, such as Intel's 82559 or the like. Other types of network connection devices can also be used. For example, the invention can use Gbit Ethernet 1394, and USB 2.0. The network controller couples to a hub <b>233</b> in the console, which includes shared peripheral system <b>201</b>.

Also shown is the second ACM <b>205</b>. The second ACM has the same or similar components as the first ACM. Here, like reference numerals have been used for easy cross-referencing, but is not intended to be limiting. In some embodiments, the secondary ACM is not connected to the PCI bus in the peripheral console directly. The secondary ACM <b>219</b> accesses peripheral devices controlled by the primary ACM through the Ethernet connection to the primary ACM, e.g. CD-ROM, or PCI modem. The implementation is not restricted to Ethernet serial communication and can use other high-speed serial communication such as USB 2.0, and 1394. The Ethernet hub is coupled to an external output port <b>235</b>, which connects to an external network.

The primary hard disk drive in each ACM can be accessed by the other ACM as sharable hard drive through the Ethernet connection. This allows the easy sharing of files between the two independent computer modules. The Ethernet Hub Controller provides the high-speed communication function between the two computer modules. Ethernet data bandwidth of 100 Mbit/sec allows fast data communication between the two computer modules. The secondary ACM access peripheral devices of the primary ACM through the network connection provided by Ethernet link. The operating system, e.g. Windows98, provides the sharing of resources between the two ACMs. In some embodiments, critical data in one ACM can be backup into the other ACM.

The Ethernet hub also couples to PCI bus <b>239</b>, which connects to PCI devices <b>241</b>, <b>243</b>, e.g., modem, SCSI controller. A flash memory <b>242</b> can also be coupled to the PCI bus. The flash memory can store passwords and security information, such as those implementations described in U.S. Ser. No. 09/183,493, which is commonly owned, and hereby incorporated by reference. The hub <b>233</b> also couples to an I/O control <b>237</b>, which connects to keyboard/mouse switch <b>245</b>, which couples to keyboard/mouse <b>247</b>. Optionally, the keyboard/mouse switch also couples to a second keyboard/house <b>259</b> via PS2 or USB signal line <b>251</b>. The keyboard/mouse switch has at least a first state and a second state, which allow operation of respectively multiple keyboards or a single keyboard. The switch also couples to each I/O controller <b>221</b> in each ACM via lines <b>253</b>, <b>255</b>. The I/O control <b>237</b> also couples to an RGB switch <b>257</b>, which allows video signals to pass to the first monitor <b>259</b>. Alternatively, the RGB switch couples to a second monitor <b>261</b>. The RGB switch includes analog video switches such as MAXIM's MAX4545.

The peripheral system <b>201</b> also has an independent power supply <b>231</b> for each ACM. Each power supply provides power to each ACM. As merely an example, the power supply is a MICRO ATX 150W made by ENLIGHT, but can be others. The power supply is connected or coupled to each ACM through a separate line, for example. The independent power supply allows for independent operation of each ACM in some embodiments.

The above embodiments are described generally in terms of hardware and software. It will be recognized, however, that the functionality of the hardware can be further combined or even separated. The functionality of the software can also be further combined or even separated. Hardware can be replaced, at times, with software. Software can be replaced, at times, with hardware. Accordingly, the present embodiments should not be construed as limiting the scope of the claims here. One of ordinary skill in the art would recognize other variations, modifications, and alternatives.

<figref idref="DRAWINGS">FIG. 3</figref> is a simplified block diagram <b>300</b> of a computer system according to an alternative embodiment of the present invention. This diagram is merely an example which should not limit the scope of the claims herein. One of ordinary skill in the art would recognizes many other variations, modifications, and alternatives. Like reference numerals are used in this Fig. as the previous Figs. for easy referencing, but are not intended to be limiting. As shown, each ACM includes common elements as the previous Fig. A primary ACM <b>203</b> is connected to PCI peripheral devices in the peripheral console through the PCI bus <b>225</b> that passes through the connection between primary ACM <b>203</b> and peripheral console <b>201</b>. As shown, ACM has a CPU module <b>207</b> coupled to the PCI bus through a North Bridge <b>211</b>.

The CPU module can use a suitable microprocessing unit, microcontroller, digital signal processor, and the like. In a specific embodiment, the CPU module uses, for example, a 400 MHz Pentium II microprocessor module from Intel Corporation and like microprocessors from AMD Corporation, Cyrix Corporation (now National Semiconductor Corporation), and others. In other aspects, the microprocessor can be one such as the Compaq Computer Corporation Alpha Chip, Apple Computer Corporation PowerPC G3 processor, and the like. Further, higher speed processors are contemplated in other embodiments as technology increases in the future.

In the CPU module, peripheral controller <b>213</b> is coupled to BIOS/flash memory <b>217</b>. Additionally, the peripheral controller is coupled to a clock control logic, a configuration signal, and a peripheral bus. The ACM has the hard drive module <b>215</b>. Among other elements, the ACM includes north bridge <b>215</b>, graphics subsystem <b>223</b> (e.g., graphics accelerator, graphics memory), an IDE controller, and other components. Adjacent to and in parallel alignment with the hard drive module <b>215</b> is the PCI bus. In a specific embodiment, North Bridge unit <b>211</b> often couples to a computer memory <b>209</b>, to the graphics subsystem, and to the peripheral controller via the PCI bus. Graphics subsystem typically couples to a graphics memory, and other elements. IDE controller generally supports and provides timing signals necessary for the IDE bus. In the present embodiment, the IDE controller is embodied as part of a P114XE controller from Intel, for example. Other types of buses than IDE are contemplated, for example EIDE, SCSI, 1394, and the like in alternative embodiments of the present invention.

The hard drive module or mass storage unit <b>215</b> typically includes a computer operating system, application software program files, data files, and the like. In a specific embodiment, the computer operating system may be the Windows98 operating system from Microsoft Corporation of Redmond Wash. Other operating systems, such as WindowsNT, MacOS8, Unix, and the like are also contemplated in alternative embodiments of the present invention. Further, some typical application software programs can include Office98 by Microsoft Corporation, Corel Perfect Suite by Corel, and others. Hard disk module <b>215</b> includes a hard disk drive. The hard disk drive, however, can also be replaced by removable hard disk drives, read/write CD ROMs, flash memory, floppy disk drives, and the like. A small form factor, for example 2.5″, is currently contemplated, however, other form factors, such as PC card, and the like are also contemplated. Mass storage unit <b>240</b> may also support other interfaces than IDE.

Among other features, the computer system includes an ACM with security protection.

The ACM also has a network controller, which can be coupled to a serial port <b>302</b>, which is coupled to the PCI bus in the ACM. The serial port is coupled to the peripheral console through a serial controller <b>301</b> in the serial console. The serial controller is connected to PCI bus <b>239</b>. The serial controller is also coupled to a serial hub controller <b>303</b>, which is coupled to the PCI bus and a second ACM. In a specific embodiment, a receptacle board <b>310</b> is added to connect to the second ACM. The purpose of the receptacle board is to allow a cable connection <b>307</b> to the peripheral board <b>300</b>. The cable connection is possible because the signals needed to connect to the peripheral board can be limited to video, I/O, serial communication, and power. The serial communication controller can be placed on the receptacle board and not in the ACM. As shown, the serial bus controller couples to the PCI bus. The receptacle board also couples to power, graphics subsystem, I/O controller, and other elements, which may be on a common bus. The overall operation of the present configuration is similar to the previous one except it operates in serial communication mode.

The Dual ACM system can support different usage models:

1. One user using both ACMs concurrently with 1 or 2 monitors, and a common keyboard/mouse.

2. Two users using the two separate ACMs at the same time with separate monitors and keyboard/mouse. The 2 users share peripherals, e.g., printer, CDROM, and others. The two users share external networking.

To support 1 monitor for both ACMs, a video switch in the peripheral console is used to switch between the video outputs of the two ACMs. The system can be set to support either 1 monitor or 2-monitor mode. The user presses a special key on the keyboard or a special icon on the screen to switch the screen display from one ACM to the other. This same action causes the keyboard and mouse connections to switch from one ACM to the other ACM.

A dual ACM system can save space, wiring, and cost for a 2-person PC setup, with the added benefit that both PC systems can be accessed from one user site for increased system performance if the other user is not using the system. Files can be copied between the primary drive of both system and provides protection against a single ACM failure. Software needs to be developed to manage the concurrent use of two PC subsystems, the automatic sharing of selected files between the two systems, and fault tolerance.

The design with more than two computer modules can be implemented with the use of multi-port, serial communication hub controller and multi-port I/O switches. In one embodiment, a peripheral console has four computer bays for four separate computer modules. The computer modules communicate through a four port Ethernet hub. The video, keyboard, and mouse switch will cycle through the connection from each computer module to the external monitor, keyboard, and mouse with a push button sequentially. This embodiment is useful for a server that performs different functions concurrently, e.g. email, application hosting, web hosting, firewall, etc.

The above embodiments are described generally in terms of hardware and software. It will be recognized, however, that the functionality of the hardware can be further combined or even separated. The functionality of the software can also be further combined or even separated. Hardware can be replaced, at times, with software. Software can be replaced, at times, with hardware. Accordingly, the present embodiments should not be construed as limiting the scope of the claims here. One of ordinary skill in the an would recognize other variations, modifications, and alternatives.

<figref idref="DRAWINGS">FIG. 4</figref> is a simplified diagram of a method according to an embodiment of the present invention. This diagram is merely an example which should not limit the scope of the claims herein. One of ordinary skill in the art would recognize many other variations, modifications, and alternatives. The present diagram illustrates an automatic file backup procedure from one computer module to the other. As shown, a user selects (step <b>401</b>) a certain file in one of the computer module for automatic backup. Next, the method determines if another module is available, step <b>403</b>. If so, the method in the originating module requests the other computer module to create (step <b>405</b>) backup file. Alternatively, the method alerts the user of the missing or malfunctioning module, step <b>429</b>. The method then has the user try later <b>431</b>, once the missing or malfunctioning module has been replaced or repaired. Next, the method determines if there is sufficient storage available in the other computer module for the backup files. If so, the method goes to the next step. (Alternatively, the method prompts (step <b>433</b>) a message to the user indicating that the storage is full.) In the next step, the method stores the backup file in memory of the other module. After the backup file has been successfully created (step <b>409</b>), the software in the originating ACM sets a timer to check (step <b>411</b>) for file modification via branches <b>423</b>, <b>427</b> through continue, step <b>425</b> process. If a file selected for backup has been modified (step <b>415</b>), then the file is automatically back up to the other ACM again, step <b>417</b>. Alternatively, the method returns to step <b>411</b> through branch <b>421</b>.

The above embodiments are described generally in terms of hardware and software. It will be recognized, however, that the functionality of the hardware can be further combined or even separated. The functionality of the software can also be further combined or even separated. Hardware can be replaced, at times, with software. Software can be replaced, at times, with hardware. Accordingly, the present embodiments should not be construed as limiting the scope of the claims here. One of ordinary skill in the art would recognize other variations, modifications, and alternatives.

<figref idref="DRAWINGS">FIG. 6</figref> is a block diagram of one embodiment of a computer system <b>600</b> using the interface of the present invention. Computer system <b>600</b> includes an attached computer module (ACM) <b>605</b> and a peripheral console <b>610</b>, which are described in greater detail in the application of William W. Y. Chu for “Personal Computer Peripheral Console With Attached Computer Module” filed concurrently with the present application on Sep. 8, 1998 and incorporated herein by reference. The ACM <b>605</b> and the peripheral console <b>610</b> are interfaced through an exchange interface system (XIS) bus <b>615</b>. The XIS bus <b>615</b> includes power bus <b>616</b>, video bus <b>617</b> and peripheral bus (XPBus) <b>618</b>, which is also herein referred to as an interface channel. The power bus <b>616</b> transmits power between ACM <b>605</b> and peripheral console <b>610</b>. In a preferred embodiment power bus <b>616</b> transmits power at voltage levels of 3.3 volts, 5 volts and 12 volts. Video bus <b>617</b> transmits video signals between the ACM <b>605</b> and the peripheral console <b>610</b>. In a preferred embodiment, the video bus <b>617</b> transmits analog Red Green Blue (RGB) video signals for color monitors, digital video signals (such as Video Electronics Standards Association (VESA) Plug and Display's Transition Minimized Differential Signaling (TMDS) signals for flat panel displays), and television (TV) and/or super video (S-video) signals. The XPBus <b>618</b> is coupled to host interface controller (HIC) <b>619</b> and to peripheral interface controller (PIC) <b>620</b>, which is also sometimes referred to as a bay interface controller.

In the embodiment shown in <figref idref="DRAWINGS">FIG. 6</figref>, HIC <b>619</b> is coupled to an integrated unit <b>621</b> that includes a CPU, a cache and a north bridge. In another embodiment, such as that shown in <figref idref="DRAWINGS">FIG. 7</figref>, the CPU <b>705</b> and north bridge <b>710</b> are separate rather than integrated units. In yet another embodiment, such as that shown in <figref idref="DRAWINGS">FIG. 8</figref>, the HIC and PIC are integrated with the north and south bridges, respectively, such that integrated HIC and north bridge unit <b>805</b> includes an HIC and a north bridge, while integrated PIC and south bridge unit <b>810</b> includes a PIC and a south bridge.

<figref idref="DRAWINGS">FIG. 9</figref> is a more detailed block diagram of one embodiment of an HIC <b>905</b> and a PIC <b>955</b> of the present invention. HIC <b>905</b> includes a peripheral component interconnect (PCI) bus controller <b>910</b>, an XPBus controller <b>915</b>, a phase lock loop (PLL) clock <b>920</b> and an input/output (IO) control <b>925</b>. Similarly, PIC <b>955</b> includes a PCI bus controller <b>960</b>, an XPBus controller <b>965</b>, a PLL clock <b>970</b> and an IO control <b>975</b>. PCI bus controllers <b>910</b> and <b>960</b> are coupled to the primary and secondary PCI buses <b>930</b> and <b>980</b>, respectively, and manage PCI transactions on the primary and secondary PCI buses <b>930</b> and <b>980</b>, respectively. Similarly, XPBus Controllers <b>915</b> and <b>965</b> are coupled to XPBus <b>990</b>. XPBus controller <b>915</b> drives the PCK line <b>991</b> and PD[0::3] and PCN lines <b>992</b> while XPBus controller <b>965</b> drives the PCKR lines <b>993</b>, the PDR[0::3] and PCNR lines <b>994</b> and the RESET# line <b>995</b>.

PCI bus controller <b>910</b> receives PCI clock signals from the primary PCI bus <b>930</b> and is synchronized to the PCI clock. However, as indicated in <figref idref="DRAWINGS">FIG. 9</figref>, the XPBus controller <b>915</b> is asynchronous with the PCI bus controller <b>910</b>. Instead, the XPBus controller receives a clock signal from the PLL clock <b>920</b> and is synchronized therewith. PLL clock <b>920</b> generates a clock signal independent of the PCI clock. The asynchronous operation of the PCI bus and the XPBus allows the PCI Bus to change in frequency. for example as in a power down situation, without directly affecting the XPBus clocking. In the embodiment shown in <figref idref="DRAWINGS">FIG. 9</figref>, the PLL clock <b>920</b> generates a clock signal having a frequency of 66 MHz, which is twice as large as the 33 MHz frequency of the PCI clock. (The clock signal generated by the PLL clock may have a clock speed different from, including lower than, 66 MHz. For example, in another embodiment, which is discussed in greater detail below, the PLL clock <b>920</b> generates a clock signal having a frequency of 132 MHz.)

The XPBus <b>990</b> operates at the clock speed generated by the PLL clock <b>920</b>. Therefore, PCK, the clock signal from the XPBus controller <b>915</b> to XPBus controller <b>965</b> has the same frequency as the clock signal generated by PLL clock <b>920</b>. XPBus controller <b>965</b> receives the PCK signal after it has been buffered and operates at the clock speed of PCK. The buffered version of the clock signal PCK is used to generate the clock signal PCKR, the clock signal form the XPBus controller <b>965</b> to XPBus controller <b>915</b>. Accordingly, PCKR also has the same frequency as that generated by the PLL clock <b>920</b>. The synchronous operation of PCK and PCKR provides for improved reliability in the system. In another embodiment, PCKR may be generated independently of PCK and may have a frequency different from that of PCK. It is to be noted that even when PCKR is generated from PCK, the slew between PCK and PCKR cannot be guaranteed because of the unknown cable length used for the XPBus. For a cable that is several feet long, the cable propagation delay alone can be several nano seconds.

As indicated in <figref idref="DRAWINGS">FIG. 9</figref>, PLL clock <b>970</b> is asynchronous with the XPBus controller <b>965</b>. Instead, PLL clock <b>970</b> independently generates a clock signal that is used as a PCI clock signal on the secondary PCI bus <b>980</b>. The secondary PCI bus <b>980</b> operates at the same clock speed as the primary PCI bus <b>930</b>, namely at a frequency of 33 MHz.

<figref idref="DRAWINGS">FIG. 10</figref> is a detailed block diagram of one embodiment of the HIC of the present invention. As shown in <figref idref="DRAWINGS">FIG. 10</figref>, HIC <b>1000</b> comprises bus controller <b>1010</b>, translator <b>1020</b>, transmitter <b>1030</b>, receiver <b>1040</b>, a PLL <b>1050</b>, an address/data multiplexer (A/D MUX) <b>1060</b>, a read/write controller (RD/WR Cntl) <b>1070</b>, a video serial to parallel converter <b>1080</b> and a CPU control &amp; general purpose input/output latch/driver (CPU CNTL &amp; GPIO latch/driver) <b>1090</b>.

HIC <b>1000</b> is coupled to an optional flash memory BIOS configuration unit <b>1001</b>. Flash memory unit <b>1001</b> stores basic input output system (BIOS) and PCI configuration information and supplies the BIOS and PCI configuration information to A/D MUX <b>1060</b> and RD/WR Control <b>1070</b>, which control the programming, read, and write of flash memory unit <b>1001</b>.

Bus controller <b>1010</b> is coupled to the host PCI bus. which is also referred to herein as the primary PCI bus, and manages PCI bus transactions on the host PCI bus. Bus controller <b>1010</b> includes a slave (target) unit <b>1011</b> and a master unit <b>1016</b>. Both slave unit <b>1011</b> and master unit <b>1016</b> each include two first in first out (FIFO) buffers, which are preferably asynchronous with respect to each other since the input and output of the two FIFOs in the master unit <b>1016</b> as well as the two FIFOs in the slave unit <b>1011</b> are clocked by different clocks, namely the PCI clock and the PCK. Additionally, slave unit <b>1011</b> includes encoder <b>1022</b> and decoder <b>1023</b>, while master unit <b>1016</b> includes encoder <b>1027</b> and decoder <b>1028</b>. The FIFOs <b>1012</b>, <b>1013</b>, <b>1017</b> and <b>1018</b> manage data transfers between the host PCI bus and the XPBus, which in the embodiment shown in <figref idref="DRAWINGS">FIG. 10</figref> operate at 33 MHz and 106 MHz, respectively. PCI address/data (AD) from the host PCI bus is entered into FIFOs <b>1012</b> and <b>1017</b> before they are encoded by encoders <b>1022</b> and <b>1023</b>. Encoders <b>1022</b> and <b>1023</b> format the PCI address/data bits to a form more suitable for parallel to serial conversion prior to transmittal on the XPBus. Similarly, address and data information from the receivers is decoded by decoders <b>1023</b> and <b>1028</b> to a form more suitable for transmission on the host PCI bus. Thereafter the decoded data and address information is passed through FIFOs <b>1013</b> and <b>1018</b> prior to being transferred to the host PCI bus. FIFOs <b>1012</b>, <b>1013</b>, <b>1017</b> and <b>1018</b>, allow bus controller <b>1010</b> to handle posted and delayed PCI transactions and to provide deep buffering to store PCI transactions.

Bus controller <b>1010</b> also comprises slave read/write control (RD/WR Cntl) <b>1014</b> and master read/write control (RD/WR Cntl) <b>1015</b>. RD/WR controls <b>1014</b> and <b>1015</b> are involved in the transfer of PCI control signals between bus controller <b>1010</b> and the host PCI bus.

Bus controller <b>1010</b> is coupled to translator <b>1020</b>. Translator <b>1020</b> comprises encoders <b>1022</b> and <b>1027</b>, decoders <b>1023</b> and <b>1028</b>, control decoder &amp; separate data path unit <b>1024</b> and control encoder &amp; merge data path unit <b>1025</b>. As discussed above encoders <b>1022</b> and <b>1027</b> are part of slave data unit <b>1011</b> and master data unit <b>1016</b>, respectively, receive PCI address and data information from FIFOs <b>1012</b> and <b>1017</b>, respectively, and encode the PCI address and data information into a form more suitable for parallel to serial conversion prior to transmittal on the XPBus. Similarly, decoders <b>1023</b> and <b>1028</b> are part of slave data unit <b>1011</b> and master data unit <b>1016</b>, respectively, and format address and data information from receiver <b>1040</b> into a form more suitable for transmission on the host PCI bus. Control encoder &amp; merge data path unit <b>1025</b> receives PCI control signals from the slave RD/WR control <b>1014</b> and master RD/WR control <b>1015</b>. Additionally. control encoder &amp; merge data path unit <b>1025</b> receives control signals from CPU CNTL &amp; GPIO latch/driver <b>1090</b>, which is coupled to the CPU and north bridge (not shown in <figref idref="DRAWINGS">FIG. 10</figref>). Control encoder &amp; merge data path unit <b>1025</b> encodes PCI control signals as well as CPU control signals and north bridge signals into control bits, merges these encoded control bits and transmits the merged control bits to transmitter <b>1030</b>, which then transmits the control bits on the data lines PD<b>0</b> to PD<b>3</b> and control line PCN of the XPBus. Examples of control signals include PCI control signals and CPU control signals. A specific example of a control signal is FRAME# used in PCI buses. A control bit, on the other hand is a data bit that represents a control signal. Control decoder &amp; separate data path unit <b>1024</b> receives control bits from receiver <b>1040</b> which receives control bits on data lines PDR<b>0</b> to PDR<b>3</b> and control line PCNR of the XPBus. Control decoder &amp; separate data path unit <b>1024</b> separates the control bits it receives from receiver <b>1040</b> into PCI control signals, CPU control signals and north bridge signals, and decodes the control bits into PCI control signals, CPU control signals, and north bridge signals all of which meet the relevant timing constraints.

Transmitter <b>1030</b> receives multiplexed parallel address/data (AID) bits and control bits from translator <b>1020</b> on the AD[31::0] out and the CNTL out lines, respectively. Transmitter <b>1030</b> also receives a clock signal from PLL <b>1050</b>. PLL <b>1050</b> takes a reference input clock and generates PCK that drives the XPBus. PCK is asynchronous with the PCI clock signal and operates at 106 MHz, twice the speed of the PCI clock of 33 MHz. The higher speed is intended to accommodate at least some possible increases in the operating speed of future PCI buses. As a result of the higher speed, the XPBus may be used to interface two PCI or PCI-like buses operating at 106 MHz rather than 33 MHz or having 104 rather than 32 multiplexed address/data lines.

The multiplexed parallel A/D bits and some control bits input to transmitter <b>1030</b> are serialized by parallel to serial converters <b>1032</b> of transmitter <b>1030</b> into 10 bit packets. These bit packets are then output on data lines PD<b>0</b> to PD<b>3</b> of the XPBus. Other control bits are serialized by parallel to serial converter <b>1033</b> into 10 bit packets and send out on control line PCN of the XPBus.

A 10× multiplier <b>1031</b> receives PCK, multiplies it by a factor of 10 and feeds a clock signal 10 times greater than PCK into the parallel to serial converters <b>1032</b> and <b>1033</b>. The parallel to serial converters <b>1032</b> and <b>1033</b> perform bit shifting at 10 times the PCK rate to serialize the parallel bits into 10 bit packets. As the parallel to serial converters <b>1032</b> and <b>1033</b> shift bits at 10 times the PCK rate, the bit rate for the serial bits output by the parallel to serial converters is 10 times higher than PCK rate, i.e., 1060 MHz. However, the rate at which data packets are transmitted on the XPBus is the same as the PCK rate, i.e., 106 MHz. As the PCI buses operate at a clock and bit rate of 33 MHz, the XPBus has a clock rate that is twice as large and a bit rate per bit line (channel) that is 100 times as large as that of the PCI buses which it interfaces.

Receiver <b>1040</b> receives serial bit packets on data lines PDR<b>0</b> to PDR<b>3</b> and control line PCNR. Receiver <b>1040</b> also receives PCKR on the XPBus as well as the clock signal PCK from PLL <b>1050</b>. The synchronizer (SYNC) <b>1044</b> of receiver <b>1040</b> synchronizes the clock signal PCKR to the locally generated clock signal, PCK, in order to capture the bits received from the XPBus into PCK clock timing.

Serial to parallel converters <b>1042</b> convert the serial bit packets received on lines PDR<b>0</b> to PDR<b>3</b> into parallel address/data and control bits that are sent to decoders <b>1023</b> and <b>1028</b> and control decoder and separate data path unit <b>1024</b>, respectively. Serial to parallel converter <b>1043</b> receives control bit packets from control line PCNR, converts them to parallel control bits and sends the parallel control bits to control decoder &amp; separate data path <b>1024</b>.

A 10× multiplier <b>1041</b> receives PCKR, multiplies it by a factor of 10 and feeds a clock signal 10 times greater than PCKR into the serial to parallel converters <b>1042</b> and <b>1043</b>. Because the bits on PDR<b>0</b> to PDR<b>3</b> and PCNR are transmitted at a bit rate of 10 times the PCKR rate, the serial to parallel converters <b>1042</b> and <b>1043</b> perform bit shifting at 10 times the PCKR rate to convert the 10 bit packets into parallel bits. It is to be noted that the rate at which bit packets are transmitted on the XPBus is the same as the PCKR rate, i.e., 106 MHz. The parallel data and control bits are thereafter sent to decoders <b>1023</b> and <b>1028</b> by way of the AD[3::0] in line and to control decoder &amp; separate data path unit <b>1024</b> by way of CNTL in lines, respectively.

Reset control unit <b>1045</b> of HIC <b>1000</b> receives the signal RESET#, which is an independent system reset signal, on the reset line RESET#. Reset control unit <b>1045</b> then transmits the reset signal to the CPU CNTL &amp; GPIO latch/driver unit <b>1090</b>.

As may be noted from the above, the 32 line host and secondary PCI buses are interfaced by 10 XPBus lines (PD<b>0</b>, PD<b>1</b>, PD<b>2</b>, PD<b>3</b>, PCN, PDR<b>0</b>, PDR<b>1</b>, PDR<b>2</b>, PDR<b>3</b>, PCNR). Therefore, the interface channel, XPBus, of the present invention uses fewer lines than are contained in either of the buses which it interfaces, namely the PCI buses. XPBus is able to interface such PCI buses without backup delays because the XPBus operates at a clock rate and a per line (channel) bit rate that are higher than those of the PCI buses.

In addition to receiving a reset signal, the CPU CNTL &amp; GPIO latch/driver <b>1090</b> is responsible for latching input signals from the CPU and north bridge and sending the signals to the translator. It also takes decoded signals from the control decoder &amp; separate data path unit <b>1024</b> and drives the appropriate signals for the CPU and north bridge.

In the embodiment shown in <figref idref="DRAWINGS">FIG. 10</figref>, video serial to parallel converter <b>1080</b> is included in HIC <b>1000</b>. In another embodiment, video serial to parallel converter <b>1080</b> may be a separate unit from the HIC <b>1000</b>. Video serial to parallel converter <b>1080</b> receives serial video data on line VPD and a video clock signal VPCK from line VPCK of video bus <b>1081</b>. It then converts the serial video data into 16 bit parallel video port data and the appropriate video port control signals, which it transmits to the graphics controller (not shown in <figref idref="DRAWINGS">FIG. 10</figref>) on the video port data [0::15] and video port control lines, respectively.

HIC <b>1000</b> handles the PCI bus control signals and control bits from the XPBus representing PCI control signals in the following ways:

1. HIC <b>1000</b> buffers clocked control signals from the host PCI bus, encodes them into control bits and sends the encoded control bits to the XPBus;

2. HIC <b>1000</b> manages the signal locally; and

3. HIC <b>1000</b> receives control bits from XPBus, translates the control bits into PCI control signals and sends the PCI control signals to the host PCI bus.

<figref idref="DRAWINGS">FIG. 11</figref> is a detailed block diagram of one embodiment of the PIC of the present invention. PIC <b>1100</b> is nearly identical to HIC <b>1000</b> in its function, except that HIC <b>1000</b> interfaces the host PCI bus to the XPBus while PIC <b>1100</b> interfaces the secondary PCI bus to the XPBus. Similarly, the components in PIC <b>1100</b> serve the same function as their corresponding components in HIC <b>1000</b>. Reference numbers for components in PIC <b>1100</b> have been selected such that a component in PIC <b>1100</b> and its corresponding component in HIC <b>1000</b> have reference numbers that differ by 500 and have the same two least significant digits. Thus for example, the bus controller in PIC <b>1100</b> is referenced as bus controller <b>1110</b> while the bus controller in HIC <b>1000</b> is referenced as bus controller <b>1010</b>. As many of the elements in PIC <b>1100</b> serve the same functions as those served by their corresponding elements in HIC <b>1000</b> and as the functions of the corresponding elements in HIC <b>1000</b> have been described in detail above, the function of elements of PIC <b>1100</b> having corresponding elements in HIC <b>1000</b> will not be further described herein. Reference may be made to the above description of <figref idref="DRAWINGS">FIG. 10</figref> for an understanding of the functions of the elements of PIC <b>1100</b> having corresponding elements in HIC <b>1000</b>.

As suggested above, there are also differences between HIC <b>1000</b> and PIC <b>1100</b>. Some of the differences between HIC <b>1000</b> and PIC <b>1100</b> include the following. First, receiver <b>1140</b> in PIC <b>1100</b>, unlike receiver <b>1040</b> in HIC <b>1000</b>, does not contain a synchronization unit. As mentioned above, the synchronization unit in HIC <b>1000</b> synchronizes the PCKR clock to the PCK clock locally generated by PLL <b>1050</b>. PIC <b>1100</b> does not locally generate a PCK clock and therefore, it does not have a locally generated PCK clock with which to synchronize the PCK clock signal that it receives from HIC <b>1000</b>. Another difference between PIC <b>1100</b> and HIC <b>1000</b> is the fact that PIC <b>1100</b> contains a video parallel to serial converter <b>1189</b> whereas HIC <b>1000</b> contains a video serial to parallel converter <b>1080</b>. Video parallel to serial converter <b>1189</b> receives 16 bit parallel video capture data and video control signals on the Video Port Data [0::15] and Video Port Control lines, respectively, from the video capture circuit (not shown in <figref idref="DRAWINGS">FIG. 11</figref>) and converts them to a serial video data stream that is transmitted on the VPD line to the HIC. The video capture circuit may be any type of video capture circuit that outputs a 16 bit parallel video capture data and video control signals. Another difference lies in the fact that PIC <b>1100</b>, unlike HIC <b>1000</b>, contains a clock doubler <b>1182</b> to double the video clock rate of the video clock signal that it receives. The doubled video clock rate is fed into video parallel to serial converter <b>1182</b> through buffer <b>1183</b> and is sent to serial to parallel converter <b>1080</b> through buffer <b>1184</b>. Additionally, reset control unit <b>1135</b> in PIC <b>1100</b> receives a reset signal from the CPU CNTL &amp; GPIO latch/driver unit <b>1190</b> and transmits the reset signal on the RESET# line to the HIC <b>1000</b> whereas reset control unit <b>1045</b> of HIC <b>1000</b> receives the reset signal and forwards it to its CPU CNTL &amp; GPIO latch/driver unit <b>1090</b> because, in the above embodiment, the reset signal RESET# is unidirectionally sent from the PIC <b>1100</b> to the HIC <b>1000</b>.

Like HIC <b>1000</b>, PIC <b>1100</b> handles the PCI bus control signals and control bits from the XPBus representing PCI control signals in the following ways:

1. PIC <b>1100</b> buffers clocked control signals from the secondary PCI bus. encodes them and sends the encoded control bits to the XPBus;

2. PIC <b>1100</b> manages the signal locally; and

3. PIC <b>1100</b> receives control bits from XPBus, translates them into PCI control signals and sends the PCI control signals to the secondary PCI bus.

PIC <b>1100</b> also supports a reference arbiter on the secondary PCI Bus to manage the PCI signals REQ# and GNT#.

<figref idref="DRAWINGS">FIG. 12</figref> is a table showing the symbols, signals, data rate and description of signals on the XPBus, where RTN indicates a ground (GND) reference. In the above tables, P&amp;D stands for plug and display and is a trademark of the Video Electronics Standards Association (VESA) for the Plug and Display standard, DDC2:SCL and DDC2:SDA stand for the VESA display data channel (DDC) standard 2 clock and data signals, respectively, SV stands for super video, V33 is 3.3 volts, and V5 is 5.0 volts. TMDS stands for Transition Minimized Differential Signaling and is a trademark of Silicon Images and refers to their Panel Link technology, which is in turn a trademark for their LVDS technology. TMDS is used herein to refer to the Panel Link technology or technologies compatible therewith.

<figref idref="DRAWINGS">FIG. 13</figref> is a table showing the information transmitted on the XPBus during two clock cycles of the XPBus in one embodiment of the present invention where 10 data bits are transmitted in each clock cycle of the XPBus. In <figref idref="DRAWINGS">FIG. 13</figref>, A<b>00</b> to A<b>31</b> represent 32 bits of PCI address A[31::0], D<b>00</b> to D<b>31</b> represent 32 bits of PCI data D[31::0], BS<b>0</b> to BS<b>3</b> represent 4 bits of bus status data indicating the status of the XPBus, CM<b>0</b># to CM<b>3</b># represent 4 bits of PCI command information, BE<b>0</b># to BE<b>3</b># represent 4 bits of PCI byte enable information, and CN<b>0</b> to CN<b>9</b> represent 10 bits of control information sent in each clock cycle. As shown in <figref idref="DRAWINGS">FIG. 13</figref>, for each of lines PD<b>0</b> to PD<b>3</b>, the 10 bit data packets contain one BS bit, one CM/BE bit, and eight AID bits. For the PCN line, the 10 bit data packet contains 10 CN bits. The first clock cycle shown in <figref idref="DRAWINGS">FIG. 13</figref> comprises an address cycle in which 4 BS bits, 4 CM bits, 32 A bits and 10 CN bits are sent. The second clock cycle comprises a data cycle in which 4 BS bits, 4 BE bits, 32 D bits and 10 CN bits are sent. The bits transmitted on lines PD<b>0</b> to PD<b>3</b> represent 32 PCI AD[31::0] signals, 4 PCI C/BE# [3::0] signals, and part of the function of PCI control signals, such as FRAME#, IRDY#, and TRDY#.

In the embodiment shown in <figref idref="DRAWINGS">FIG. 13</figref>, BS<b>0</b> to BS<b>3</b> are sent at the beginning of each clock cycle. The bus status bits indicate the following bus cycle transactions: idle, address transfer, write data transfer, read data transfer, switch XPBus direction, last data transfer, wait, and other cycles.

Bits representing signals transmitted between the CPU and South Bridge may also be sent on the lines interconnecting the HIC and PIC, such as lines PCN and PCNR. For example, CPU interface signals such as CPU interrupt (INTR), Address 20 Mask (A20M#), Non-Maskable Interrupt (NMI), System Management Interrupt (SMI#), and Stop Clock (STPCLK#), may be translated into bit information and transmitted on the XPBus between the HIC and the PIC.

<figref idref="DRAWINGS">FIG. 14</figref> is a table showing the information transmitted on the XPBus during four clock cycles of the XPBus in another embodiment of the present invention where 10 data bits are transmitted in each clock cycle of the XPBus. In this embodiment, the XPBus clock rate is twice as large as the PCI clock rate. This allows sending data and address bits every other XPBus cycle. As can be seen in <figref idref="DRAWINGS">FIG. 14</figref>, there are no address or data bits transmitted during the second or fourth XPBus clock cycle. The fact that the XPBus clock rate is higher than the PCI clock rate allows for compatibility of the XPBus with possible future expansions in the performance of PCI bus to higher data transfer and clock rates.

In the embodiment shown in <figref idref="DRAWINGS">FIG. 14</figref>, there are 18 control bits, CN<b>0</b> to CN<b>17</b>, transmitted in every two XPBus clock cycles. The first bit transmitted on the control line in each XPBus clock cycle indicates whether control bits CN<b>0</b> to CN<b>8</b> or control bits CN<b>9</b> to CN<b>17</b> will be transmitted in that cycle. A zero sent at the beginning of a cycle on the control line indicates that CN<b>0</b> to CN<b>8</b> will be transmitted during that cycle, whereas a one sent at the beginning of a cycle on the control line indicates that CN<b>9</b> to CN<b>17</b> will be transmitted during that cycle. These bits also indicate the presence or absence of data and address bits during that cycle. A zero indicates that address or data bits will be transmitted during that cycle whereas a one indicates that no address or data bits will be transmitted during that cycle.

In one embodiment, BS<b>0</b> and BS<b>1</b> are used to encode the PCI signals FRAME# and IRDY#, respectively. Additionally, in one embodiment, BS<b>2</b> and BS<b>3</b> are used to indicate the clock speed of the computer bus interface and the type of computer bus interface, respectively. For example. BS<b>2</b> value of zero may indicate that a 33 MHz PCI bus of 32 bits is used whereas a BS<b>2</b> value of one may indicate that a 66 MHz PCI bus of 32 bits is used. Similarly, a BS<b>3</b> value of zero may indicated that a PCI bus is used whereas a BS<b>3</b> value of one may indicated that another computer interface bus, such as an Institute of Electronics &amp; Electrical Engineers (IEEE) 1394 bus, is used.

<figref idref="DRAWINGS">FIG. 15</figref> is a schematic diagram of lines PCK, PD<b>0</b> to PD<b>3</b>, and PCN. These lines are unidirectional LVDS lines for transmitting clock signals and bits such as those shown in <figref idref="DRAWINGS">FIGS. 13 and 14</figref> from the HIC to the PIC. The bits on the PD<b>0</b> to PD<b>3</b> and the PCN lines are sent synchronously within every clock cycle of the PCK. Another set of lines, namely PCKR, PDR<b>0</b> to PDR<b>3</b>, and PCNR, are used to transmit clock signals and bits from the PIC to HIC. The lines used for transmitting information from the PIC to the HIC have the same structure as those shown in <figref idref="DRAWINGS">FIG. 15</figref>, except that they transmit data in a direction opposite to that in which the lines shown in <figref idref="DRAWINGS">FIG. 15</figref> transmit data. In other words they transmit information from the NC to the HIC. The bits on the PDR<b>0</b> to PDR<b>3</b> and the PCNR lines are sent synchronously within every clock cycle of the PCKR. Some of the examples of control information that may be sent in the reverse direction, i.e., on PCNR line, include a request to switch data bus direction because of a pending operation (such as read data available), a control signal change in the target requiring communication in the reverse direction, target busy, and transmission error detected.

The XPBus which includes lines PCK, PD<b>0</b> to PD<b>3</b>, PCN, PCKR, PDR<b>0</b> to PDR<b>3</b>, and PCNR, has two sets of unidirectional lines transmitting clock signals and bits in opposite directions. The first set of unidirectional lines includes PCK, PD<b>0</b> to PD<b>3</b>, and PCN. The second set of unidirectional lines includes PCKR, PDR<b>0</b> to PDR<b>3</b>, and PCNR. Each of these unidirectional set of lines is a point-to-point bus with a fixed transmitter and receiver, or in other words a fixed master and slave bus. For the first set of unidirectional lines, the HIC is a fixed transmitter/master whereas the PIC is a fixed receiver/slave. For the second set of unidirectional lines, the PIC is a fixed transmitter/master whereas the HIC is a fixed receiver/slave. The LVDS lines of XPBus, a cable friendly and remote system I/O bus, transmit fixed length data packets within a clock cycle.

The XPBus lines, PD<b>0</b> to PD<b>3</b>, PCN, PDR<b>0</b> to PDR<b>3</b> and PCNR, and the video data and clock lines, VPD and VPCK, are not limited to being LVDS lines, as they may be other forms of bit based lines. For example, in another embodiment, the XPBus lines may be IEEE 1394 lines.

It is to be noted that although each of the lines PCK, PD<b>0</b> to PD<b>3</b>, PCN, PCKR, PDR<b>0</b> to PDR<b>3</b>, PCNR, VPCK, and VPD is referred to as a line, in the singular rather than plural. each such line may contain more than one physical line. For example, in the embodiment shown in <figref idref="DRAWINGS">FIG. 23</figref>, each of lines PCK, PD<b>0</b> to PD<b>3</b> and PCN includes two physical lines between each driver and its corresponding receiver. The term line, when not directly preceded by the terms physical or conductive, is herein used interchangeably with a signal or bit channel which may consist of one or more physical lines for transmitting a signal. In the case of non-differential signal lines, generally only one physical line is used to transmit one signal. However, in the case of differential signal lines, a pair of physical lines is used to transmit one signal. For example, a bit line or bit channel in an LVDS or IEEE 1394 interface consists of a pair of physical lines which together transmit a signal.

A bit based line (i.e., a bit line) is a line for transmitting serial bits. Bit based lines typically transmit bit packets and use a serial data packet protocol. Examples of bit lines include an LVDS line, an IEEE 1394 line, and a Universal Serial Bus (USB) line.

<figref idref="DRAWINGS">FIG. 16</figref> is a table showing the names, types, number of pins dedicated to, and the description of the primary bus PCI signals. The pins represent those between the host PCI bus and the HIC.

<figref idref="DRAWINGS">FIG. 17</figref> is a block diagram of the components in one computer system employing the present invention. The computer system comprises an attached computer module (ACM), a peripheral console (PCON), and the interconnection apparatus between them. The ACM includes the central processing unit (CPU) <b>1710</b>, system memory <b>1720</b>, high performance devices <b>1750</b>, primary mass storage <b>1730</b>, and related interface and support circuitry <b>1740</b>. The PCON includes primary display <b>1810</b>, primary input <b>1820</b>, secondary mass storage <b>1750</b>, other devices <b>1860</b>, expansion slots <b>1870</b>, the primary power supply <b>1830</b>, and related interface and support circuitry <b>1840</b>. The interconnection apparatus <b>1900</b> includes circuitry to convey power and operational signals between the ACM and PCON.

Within the ACM <b>1700</b>, the CPU <b>1710</b> executes instructions and manipulates data stored in the system memory. The CPU <b>1710</b> and system memory <b>1720</b> represent the user's core computing power. The core computing power may also include high performance devices <b>1750</b> such as advanced graphics processor chips that greatly increase overall system performance and which, because of their speed, need to be located close to the CPU. The primary mass storage <b>1730</b> contains persistent copies of the operating system software, application software, configuration data, and user data. The software and data stored in the primary mass storage device represent the user's computing environment. Interface and support circuitry <b>1740</b> primarily includes interface chips and signal busses that interconnect the CPU, system memory, high performance devices, and primary mass storage. The interface and support circuitry also connects ACM-resident components with the ACM-to-PCON interconnection apparatus as needed.

Within the PCON <b>1800</b>, the primary display component <b>1810</b> may include an integrated display device or connection circuitry for an external display device. This primary display device may be, for example, an LCD, plasma, or CRT display screen used to display text and graphics to the user for interaction with the operating system and application software. The primary display component is the primary output of the computer system, i.e., the paramount vehicle by which programs executing on the CPU can communicate toward the user.

The primary input component <b>1820</b> of the PCON may include an integrated input device or connection circuitry for attachment to an external input device. The primary input may be, for example, a keyboard, touch screen, keypad, mouse, trackball, digitizing pad, or some combination thereof to enable the user to interact with the operating system and application software. The primary input component is the paramount vehicle by which programs executing on the CPU receive signals from the user.

The PCON may contain secondary mass storage <b>1850</b> to provide additional high capacity storage for data and software. Secondary mass storage may have fixed or removable media and may include, for example, devices such as diskette drives, hard disks, CD-ROM drives, DVD drives, and tape drives.

The PCON may be enhanced with additional capability through the use of integrated “Other Devices” <b>1860</b> or add-on cards inserted into the PCON's expansion slots <b>1870</b>. Examples of additional capability include sound generators, LAN connections, and modems. Interface and support circuitry <b>1840</b> primarily includes interface chips, driver chips, and signal busses that interconnect the other components within the PCON. The interface and support circuitry also connects PCON-resident components with the ACM-to-PCON interconnection apparatus as needed.

Importantly, the PCON houses the primary power supply <b>1830</b>. The primary power supply has sufficient capacity to power both the PCON and the ACM <b>1700</b> for normal operation. Note that the ACM may include a secondary “power supply” in the form, for example, of a small battery. Such a power supply would be included in the ACM to maintain, for example, a time-of-day clock, configuration settings when the ACM is not attached to a PCON, or machine state when moving an active ACM immediately from one PCON to another. The total energy stored in such a battery would, however, be insufficient to sustain operation of the CPU at its rated speed, along with the memory and primary mass storage, for more than a fraction of an hour, if the battery were able to deliver the required level of electrical current at all.

<figref idref="DRAWINGS">FIG. 18</figref> is a block diagram of an attached computing module (ACM) <b>1700</b>. The physical ACM package <b>1700</b> contains the ACM functional components <b>1701</b> and the ACM side of the ACM-to-PCON Interconnection <b>1900</b>. The ACM <b>1701</b> comprises a CPU component <b>1710</b>, a system memory component <b>1720</b>, a primary mass storage component <b>1730</b>, a high performance devices components <b>1750</b>, and an interface and support component <b>1740</b>.

The ACM side of the ACM-to-PCON Interconnection <b>1900</b> comprises a Host Interface Controller (HIC) component <b>1920</b> and an ACM connector component <b>1930</b>. The HIC <b>1920</b> and connector <b>1930</b> components couple the ACM functional components <b>1700</b> with the signals of an ACM-to-PCON interface bus <b>1910</b> used to operatively connect an ACM with a PCON. The ACM-to-PCON interface bus <b>1910</b> comprises conveyance for electrical power <b>1914</b> and signals for a peripheral bus <b>1912</b>, video <b>1916</b>, video port <b>1917</b>, and console type <b>1918</b>. The preferred ACM-to-PCON Interconnection <b>1900</b> is described in detail in a companion U.S. patent application Ser. No. 09/149,882, entitled “A Communication Channel and Interface Devices for Bridging Computer Interface Buses,” by the same inventor, filed on the same day herewith, and hereby incorporated by reference. The preferred ACM-to-PCON interconnection <b>1900</b> includes circuitry to transmit and receive parallel bus information from multiple signal paths as a serial bit stream on a single signal path. This reduces the number of physical signal paths required to traverse the interconnection <b>1900</b>. Further, employing low-voltage differential signaling (LVDS) on the bit stream data paths provides very reliable, high-speed transmission across cables. This represents a further advantage of the present invention.

The CPU component <b>1710</b> of the ACM functional circuitry <b>1701</b> of the presently described embodiment comprises a microprocessor <b>1712</b>, which is the chief component of the personal computer system, power supply connection point <b>1713</b>, and cache memory <b>1714</b> tightly coupled to the microprocessor <b>1712</b> by the CPU-to-cache bus <b>1714</b> comprising signal paths for address, data. and control information. The microprocessor <b>1712</b> of this embodiment is one of the models from the Pentium II family of processors from Intel Corporation. Microprocessor <b>1712</b> receives electrical power from power bus <b>1768</b> via connection point <b>1713</b>. Microprocessor <b>1712</b> couples to the Host Interface Controller (HIC) <b>1920</b> via CPU-to-HIC bus <b>1763</b> comprising signal paths to exchange control information such as an interrupt request. Microprocessor <b>1712</b> also couples to CPU Bridge <b>1746</b> via CPU main bus <b>1764</b> comprising signal paths for address, data, and control information.

The CPU Bridge component <b>1746</b> of the interface and support circuitry <b>1740</b> operates to couple the high speed CPU main bus <b>1764</b> to specialty buses of varying speeds and capability that connect other computer components. The CPU Bridge of the presently described embodiment incorporates memory controller circuitry, advanced graphics processor support circuitry, and a general, industry-standard PCI bus controller in a single package. A CPU Bridge <b>1746</b> such as the 82443LX PCI/AGP Controller from Intel Corporation may be used.

The system memory component <b>1720</b> of the ACM functional circuitry <b>1701</b> in the present embodiment comprises main system memory (RAM) <b>1722</b>, BIOS memory <b>1724</b>, and flash memory <b>1726</b>. The system memory <b>1720</b> is used to contain data and instructions that are directly addressable by the CPU. The RAM <b>1722</b> comprises volatile memory devices such as DRAM or SDRAM memory chips that do not retain their stored contents when power is removed. This form of memory represents the largest proportion of total system memory <b>1720</b> capacity. The BIOS memory <b>1724</b> comprises non-volatile memory devices such as ROM or EPROM memory chips that retain their stored contents regardless of the application of power and are read-only memory under normal operating conditions. The BIOS memory <b>1724</b> stores, for example, start-up instructions for the microprocessor <b>1712</b> and sets of instructions for rudimentary input/output tasks. The flash memory <b>1726</b> comprises non-volatile memory devices that retain their stored contents regardless of the application of power. Unlike the BIOS non-volatile memory, however, the stored contents of the flash memory <b>1726</b> are easily changed under normal operating conditions. The flash memory <b>1726</b> may be used to store status and configuration data, such as security identifiers or ACM specifications like the speed of the microprocessor <b>1712</b>. Some embodiments may combine the BIOS functions into the flash memory device, thus permitting BIOS contents to be rewritten, improving field upgradability.

The main system memory (RAM) <b>1722</b> is coupled to memory controller circuitry resident within the CPU Bridge <b>1746</b> via direct memory bus <b>1765</b>. The BIOS <b>1724</b> and flash memory <b>1726</b> are coupled to HIC <b>1920</b> via switched memory bus <b>1766</b>. This permits the BIOS <b>1724</b> and flash <b>1726</b> memories to be accessed by circuitry in the HIC <b>1920</b> or other circuitry connected thereto. The direct memory bus <b>1765</b> and the switch memory bus <b>1766</b> each comprises conductors to convey signals for data, address, and control information.

The primary mass storage component <b>1730</b> of the ACM functional circuitry <b>1701</b> in the present embodiment comprises a compact hard disk drive with an industry-standard, IDE interface. The hard disk drive (HDD) <b>1732</b> has a formatted storage capacity sufficient to contain an operating system for the computer, application software desired by the user, and related user configuration and operating parameter data. The HDD <b>1732</b> in the present embodiment serves as the “boot” device for the personal computer from which the operating system is loaded into RAM <b>1722</b> by the start-up program stored in the BIOS <b>1724</b>.

The present HDD <b>1732</b> has a capacity of approximately 2,000 megabytes to provide adequate storage for common software configurations and reasonable space for user data. One example of a common software configuration includes the Windows 95 operating system from Microsoft Corporation, a word processing program, a spreadsheet program, a presentation graphics program, a database program, an email program, and a web browser such as Navigator from Netscape Corporation. The hard disk <b>1732</b> stores program and data files for each software component, including files distributed by the vendor as well as files created or updated by operation of the software after it is installed. For example, a word processor program may maintain information about a user's identity and latest preferences in an operating system registry file. Or, for example, the web browser may maintain a file of the user's favorite web sites or most recently viewed web pages. An HDD with 2000 megabyte capacity is readily available in the small size of hard disk (e.g., 2.5-inch or 3.5-inch) to minimize the space required within the ACM for the primary mass storage device <b>1730</b>.

The HDD <b>1732</b> is coupled to IDE controller circuitry <b>1748</b> via IDE bus <b>1772</b>. The IDE controller circuitry <b>1748</b> is coupled to the CPU Bridge <b>1746</b> via the Host PCI bus <b>1767</b>. IDE controllers and busses, and the PCI bus are well known and understood in the industry. The above components operate together to couple the hard disk drive <b>1732</b> to the microprocessor <b>1712</b>.

The high performance devices component <b>1750</b> of the ACM functional circuitry <b>1701</b> in the present embodiment comprises an Advanced Graphics Processor (AGP) <b>1752</b>. The Model 740 Graphics Device froth Intel Corporation may be used in the present embodiment as the AGP.

Increases in computer screen size, graphics resolution, color depth, and visual motion frame rates, used by operating system and application software alike, have increased the computing power required to generate and maintain computer screen displays. An AGP removes a substantial portion of the graphics computing burden from the CPU to the specialized high-performance processor, but a high level of interaction between the CPU and the specialized processor is nonetheless required. To maximize the effective contribution of having a specialized processor in the presently described embodiment, the AGP <b>1752</b> is located in the ACM <b>1700</b>, where it is in close proximity to the microprocessor <b>1712</b>. The AGP <b>1752</b> is coupled to the microprocessor <b>1712</b> via the advanced graphics port bus <b>1773</b> of the CPU Bridge <b>1746</b>. The visual display signal generated by the AGP are conveyed toward actual display devices at the peripheral console (PCON) via video signal bus <b>1770</b>. Video information from a source external to the ACM and appearing as video port signals <b>1917</b> may be conveyed to the AGP <b>1752</b> via video port signal path <b>1771</b>.

Other types of high performance components may be included in different ACM configurations. For example, an interface to an extremely high speed data communication facility may be desirable in some future computer where CPU-to-network interaction is of comparable intensity to today's CPU-to-graphics interaction. Because such high performance components tend to be high in cost, their inclusion in the ACM is desirable. Inclusion of high cost, high performance components in the ACM concentrates a user's core computing power and environment in a portable package. This represents a further advantage of the invention.

The interface and support component <b>1740</b> of the ACM functional circuitry <b>1701</b> in the present embodiment comprises circuitry for power regulation <b>1742</b>, clocking <b>1744</b>, CPU Bridge <b>1746</b>, IDE controller <b>1748</b>, and signal conveyance paths <b>1761</b>-<b>1774</b>. The CPU Bridge <b>1746</b> couples the CPU component <b>1710</b> of the ACM <b>1700</b> with the other components of the ACM <b>1720</b>-<b>1750</b> and the CPU-to-PCON Interconnection <b>1900</b>. The CPU Bridge <b>1746</b> and IDE controller <b>1748</b> have already been discussed. Power regulation circuitry <b>1742</b> receives electrical power via the electrical power conduction path <b>1914</b> of the CPU-to-PCON Interconnection <b>1900</b>, conditions and distributes it to the other circuitry in the ACM using power distribution bus <b>1768</b>. Such regulation and distribution is well known and understood in the art.

Clocking circuitry <b>1744</b> generates clock signals for distribution to other components within the ACM <b>1700</b> that require a timing and synchronization clock source. The CPU <b>1710</b> is one such component. Often, the total power dissipated by a CPU is directly proportional to the frequency of its main clock signal. The presently described embodiment of the ACM <b>1700</b> includes circuitry that can vary the frequency of the main CPU clock signal conveyed to the CPU via signal path <b>1762</b>, in response to a signal received from the host interface controller (HIC) <b>1920</b> via signal path <b>1761</b>. The generation and variable frequency control of clocking signals is well understood in the art. By varying the frequency, the power consumption of the CPU (and thus the entire ACM) can be varied.

The variable clock rate generation may be exploited to match the CPU power consumption to the available electrical power. Circuitry in the host interface controller (HIC) <b>1920</b> of the presently described embodiment adjusts the frequency control signal sent via signal path <b>1761</b> to the clocking circuitry <b>1744</b>, based on the “console type” information signal <b>1918</b> conveyed from the peripheral console (PCON) by the CPU-to-PCON interconnection <b>1900</b>. In this arrangement, the console type signal originating from a desktop PCON would result in the generation of a maximum speed CPU clock. The desktop PCON, presumably has unlimited power from an electrical wall outlet and does not need to sacrifice speed for power conservation. The console type signal originating from a notebook PCON would, however, result in the generation of a CPU clock speed reduced from the maximum in order to conserve battery power and extend the duration of computer operation obtained from the energy stored in the battery. The console type signal originating from a notepad PCON would result in the generation of a CPU clock speed reduced further yet, the notepad PCON presumably having smaller batteries than the notebook PCON. Inclusion of control signals and circuitry to effect a CPU clock signal varying in frequency according to characteristics of the PCON to which the ACM is connected facilitates the movement of the user's core computing power and environment to different work settings, which is a further advantage of the present invention.

<figref idref="DRAWINGS">FIG. 19</figref> illustrates an external view of one embodiment of an ACM. The case <b>2010</b> of the ACM <b>1700</b> is generally rectangular in shape, preferably constructed of a strong, lightweight, rigid material that will protect the internal components from mechanical and environmental exposure. Plastics may readily be used to construct the case <b>2010</b>. The case <b>2010</b> completely surrounds the internal components, being generally an 8-sided box. <figref idref="DRAWINGS">FIG. 19</figref> shows the top <b>2012</b>, right <b>2014</b>, and rear <b>2016</b> surfaces of the ACM case <b>2010</b>. Rear edges <b>2018</b> of the case joining the rear surface <b>2016</b> with its adjoining surfaces may be beveled or rounded to facilitate insertion of the ACM <b>1700</b> into the computer bay of the PCON. Notches <b>2040</b> may be formed by projecting small surfaces inward from otherwise generally flat surfaces of the ACM case <b>2010</b>. The notches <b>2040</b> may be used to engage with mechanical devices mounted in and about a computer bay. Such mechanical devices can be employed to secure the ACM into position within a computer bay for reliability and security. Openings <b>2017</b> are formed into the rear surface <b>2016</b> of the ACM case <b>2010</b> through which to project connectors <b>1930</b><i>a </i>and <b>1930</b><i>b</i>. In one embodiment the case <b>2010</b> is approximately 5.75 inches wide by 6.5 inches deep by 1.6 inches high.

Connectors <b>1930</b><i>a </i>and <b>1930</b><i>b </i>are part of the ACM-to-PCON Interconnection as described earlier in reference to <figref idref="DRAWINGS">FIGS. 3 and 4</figref>. When the ACM <b>1700</b> is inserted into the computer bay of a peripheral console (PCON), connectors <b>1930</b><i>a </i>and <b>1930</b><i>b </i>mate with corresponding connectors located at the rear of the computer bay to electrically couple the ACM with the PCON containing the computer bay. Details concerning the ACM-to-PCON Interconnection can be found in the U.S. patent application entitled “A Communication Channel and Interface Devices for Bridging Computer Interface Buses,” already incorporated herein by reference. The connectors <b>1930</b><i>a </i>and <b>1930</b><i>b </i>used in one embodiment are connectors complying with the Device Bay industry standard as documented in “Device Bay Interface Specification,” revision 0.85, Feb. 6, 1998. Such connectors have specifically been designed to stand up to the rigors of repeated insertion and withdrawal.

Cooling plate <b>2030</b> forms part of the top surface <b>2012</b> of ACM <b>1700</b>. The cooling plate <b>2030</b> may be mounted to, or project through an opening formed in, case <b>2010</b>. Similarly, electromagnetic interference (EMI)/electrostatic discharge (ESD) rounding plate <b>2032</b> forms part of the right surface <b>2014</b> of ACM <b>1700</b>. The grounding plate <b>2032</b> may be mounted to, or project through an opening formed in, case <b>2010</b>. Cooling plate <b>2030</b> and grounding plate <b>2032</b> compressively mate with counterparts when the ACM is fully inserted into the computer bay. The counterparts located along the boundaries of the computer bay conduct dangerous heat and electrical charges away from the ACM. Inside the ACM, cooling plate <b>2030</b> thermally couples to heat-sensitive components such as CPU <b>1710</b> by methods well known in the art. Similarly, grounding plate <b>2032</b> electrically couples to EMI/ESD-sensitive components, such as a microprocessor, by methods well known in the art.

LCD display <b>2050</b> forms part of the right surface <b>2014</b> of ACM <b>1700</b>. The LCD display may be mounted to, or project through an opening formed in, case <b>2010</b>. The LCD display may contain indicators about the status of the ACM. Such indicators may display, for example, the time-of-day from a time-of-day clock contained within the ACM, or the amount of charge remaining in an ACM-resident battery, or certain configuration options recorded in flash memory. The LCD display <b>2050</b> provides display capability for a limited amount of information, most useful when the ACM is separated from a PCON (and is thus separated from a full-capability, primary display device).

<figref idref="DRAWINGS">FIG. 19</figref><i>b </i>illustrates one possible embodiment of a computer bay. A computer bay <b>1890</b> acts as a receptacle for lodging an ACM (such as the one shown in <figref idref="DRAWINGS">FIG. 20</figref>) within a desktop PCON. The illustrated computer bay <b>1890</b> provides an ACM with housing and with signal flow, electrical grounding, heat transfer, and mechanical connections. While many physical arrangements between the ACM and PCON are possible, the use of an enclosed computer bay as the one illustrated in <figref idref="DRAWINGS">FIG. 20</figref><i>b </i>offers many advantages. For example, the illustrated computer bay <b>1890</b> provides physical protection for the ACM. The computer bay may also be easily incorporated into industry standard form factors used in the manufacture of desktop personal computers (e.g., the ACM and associated computer bay could be designed to fit within the volume occupied by a standard-size disk drive).

The computer bay <b>1890</b> appearing in <figref idref="DRAWINGS">FIG. 20</figref><i>b </i>is shown mounted within the confines of PCON case <b>1802</b>. The computer bay <b>1890</b> comprises frame <b>1891</b> and signal flow, grounding, cooling, and locking components as described below. Mounting flanges <b>1898</b> of frame <b>1891</b> may be used to attach the computer bay <b>1890</b> to the PCON structure. The computer bay <b>1890</b> is prominently defined by frame <b>1891</b> generally forming a cavity in which to lodge an ACM. As such, the interior cavity formed by frame <b>1891</b> closely approximates the exterior dimensions of a compatible ACM. The top <b>1893</b>, right <b>1894</b>, and rear <b>1895</b> sides of the computer bay frame <b>1891</b> are visible. The computer bay frame <b>1891</b> also includes substantial bottom and left sides which are not shown. The front side of the frame <b>1891</b> (not shown) is open to allow the insertion of the ACM. Frame <b>1891</b> is constructed of metal for strength and to facilitate the conductance of heat and undesired electrical currents away from the ACM.

In the presently described embodiment, the weight of an inserted ACM is largely borne by the bottom side (not shown) of computer bay frame <b>1891</b>. Alternative embodiments are possible where, for example, the weight of the ACM is borne by rails running longitudinally down the right and left sides of the computer bay cavity that engage corresponding grooves running longitudinally down the right and left sides of an ACM.

<figref idref="DRAWINGS">FIG. 20</figref> illustrates the internal component layout for one embodiment of an ACM. All components are contained within the confines of the ACM case <b>510</b>, except for connectors <b>1930</b><i>a </i>and <b>1930</b><i>b </i>which extend from the rear of the ACM <b>1700</b> to engage mating connectors (not shown) that will couple the ACM circuitry with the PCON circuitry. Main circuit board <b>2110</b> provides electrical connections for circuitry within the ACM and mounting for many of its components <b>1724</b>, <b>1722</b>, <b>17221</b>, <b>1752</b>, <b>1742</b>, <b>1748</b>, <b>1920</b>, and <b>1930</b>. The fabrication and use of such circuits boards is well known and understood in the art. Connector <b>2122</b> is also mounted on main circuit board <b>2110</b> and mates with mobile processor module <b>2120</b>. Mobile processor module <b>2120</b> represents a form of packaging for a microprocessor and related components. The illustrated mobile processor module <b>2120</b> is a self-contained unit that includes a microprocessor <b>1712</b>, CPU cache <b>1714</b>, and CPU bridge <b>1746</b> operatively interconnected by the manufacturer. An example of one such module is the Pentium Processor with MMX Technology Mobile Module from Intel Corporation (order number 24 3515-001, September 1997). One skilled in the art recognizes that discrete microprocessor, cache, and bridge could have been employed and mounted directly to the main circuit board.

The mobile processor module <b>2120</b> blocks the view, from the top, of the system BIOS <b>1724</b>. Similarly, hard disk drive <b>1732</b> hides RAM memory <b>1722</b>, the high performance graphics processor <b>1752</b>, the host interface controller <b>1920</b>, and flash memory <b>1726</b>. Memory upgrade socket <b>2130</b> remains exposed to facilitate installation of additional RAM memory <b>1722</b>. Power regulator <b>1742</b>, like the memory upgrade socket, enjoys a generous amount of overhead clearance to accommodate its vertical size. The area including IDE controller <b>1748</b> also enjoys overhead clearance to facilitate a cable connection with the hard disk drive <b>1732</b>.

The functional interconnection and operation of components contained within the ACM and depicted in <figref idref="DRAWINGS">FIG. 20</figref> has already been described in relation to <figref idref="DRAWINGS">FIG. 18</figref> for like numbered items appearing therein.

<figref idref="DRAWINGS">FIG. 21</figref> is a block diagram of a peripheral console (PCON). A peripheral console couples with an ACM to form an operating personal computer system. The peripheral console (PCON) supplies an ACM with primary input, display, and power supply; the ACM supplies the core computing power and environment of the user. In the presently described embodiment the physical PCON package <b>200</b> contains the PCON functional components <b>1801</b> and the PCON side of the ACM-to-PCON Interconnection <b>1900</b>. The PCON functional components <b>1801</b> comprise primary display <b>1810</b>, a primary input <b>1820</b>, a primary power supply <b>1830</b>, interface and support <b>1840</b>, secondary mass storage <b>1850</b>, other devices <b>1860</b>, and expansion slots <b>1870</b>.

The PCON side of the ACM-to-PCON Interconnection <b>1900</b> comprises a Peripheral Interface Controller (PIC) component <b>1940</b>, a PCON connector component <b>1950</b>, console-type component <b>1942</b>, and flash memory device <b>1948</b>. The PIC <b>1940</b> and connector <b>1950</b> components couple the PCON functional components <b>1801</b> with the signals of an ACM-to-PCON interface bus <b>1910</b> used to operatively connect an ACM with a PCON. The ACM-to-PCON interface bus <b>1910</b> comprises conveyance for electrical power <b>1914</b> and signals for a peripheral bus <b>1912</b>, video <b>1916</b>, video port <b>1917</b>, and console-type <b>1918</b>. The preferred ACM-to-PCON Interconnection <b>1900</b> is described in detail in the U.S. patent application entitled “A Communication Channel and Interface Devices for Bridging Computer Interface Buses,” already incorporated herein by reference.

Connector component <b>1950</b> may be selected to mate directly with the connector component <b>1930</b> of an ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>). Alternatively, connector component <b>1950</b> may be selected to mate with, for example, the connector on one end of a cable intervening between the PCON and an ACM in a particular embodiment, such as cable <b>1964</b> shown in <figref idref="DRAWINGS">FIG. 19B</figref>. The ACM-to-PCON interconnection described in the aforementioned companion patent application has the advantage of providing reliable signal conveyance across low cost cables.

Flash memory device <b>1948</b> provides non-volatile storage. This storage may be accessible to devices in both the ACM and the PCON, including the host interface controller and the peripheral interface controller to which it is connected. As such, flash memory <b>1948</b> may be used to store configuration and security data to facilitate an intelligent mating between an ACM and a PCON that needs no participation of the CPU.

The primary display component <b>1810</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises integrated display panel <b>1812</b> and video connector <b>1813</b>. Integrated display panel <b>1812</b> is a color LCD display panel having a resolution of 640 horizontal by 480 vertical pixels. 640-by-480 resolution is popularly considered to be the minimum screen size to make practical use of the application software in widespread use today. One skilled in the an recognizes that the type and resolution of the display can vary greatly from embodiment to embodiment, depending on factors such as cost and intended application. Any display device may be used, without departing from the scope and spirit of the invention, that provides principal visual output to the computer user for operating system and application software executing in its customary and intended fashion using the CPU component (<b>1710</b> of <figref idref="DRAWINGS">FIG. 17</figref>) of an ACM presently coupled to PCON <b>1800</b>.

Integrated display panel <b>1812</b> is coupled to video signal bus <b>1849</b> and displays a screen image in response to video signals presented on bus <b>1849</b>. Certain pins of connector <b>1950</b> receive video output signals <b>1916</b> of the ACM-to-PCON interface bus <b>1910</b> from a mated connector that is coupled to an ACM. These certain pins of connector <b>1950</b> couple to video signal bus <b>1849</b> which conveys the video output signals <b>316</b> throughout the PCON <b>1800</b> as needed. Video connector <b>1813</b> is exposed at the exterior of PCON <b>1800</b> and couples to video signal bus <b>1849</b>. Connector <b>1813</b> permits easy attachment of an external display device that is compatible with the signals carried by bus <b>1849</b>, such as a CRT monitor (not shown). The external display device may be used in addition, or as an alternative, to integrated display panel <b>1812</b>.

The isolation of the relatively heavy and sizable primary display <b>1810</b> from the core computing power and user environment contained within an ACM represents a further advantage of the present invention.

The primary input component <b>1820</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises keyboard interface circuitry <b>1822</b>, keyboard connector <b>1823</b>, pointer interface circuitry <b>1824</b>, and pointer connector <b>1825</b>. Keyboard interface circuitry <b>1822</b> and pointer interface circuitry <b>1824</b> connect to ISA bus <b>1845</b> and are thereby coupled to the CPU component (<b>1710</b> of <figref idref="DRAWINGS">FIG. 17</figref>) of any ACM attached to PCON <b>1800</b>. Keyboard interface circuitry <b>1822</b> interfaces a standard computer keyboard (not shown), attached at connector <b>1823</b>, to ISA bus <b>1845</b>. Pointer interface circuitry <b>1822</b> interfaces a standard computer pointing device (not shown), such as a computer mouse attached at connector <b>1825</b>, to ISA bus <b>1845</b>. Computer keyboards, pointing devices, connectors <b>1823</b>. <b>1825</b>. keyboard interface circuitry <b>1822</b>, and pointer interface circuitry <b>1824</b> are well known in the art. The isolation of the relatively heavy and sizable primary input devices <b>1820</b> from the core computing power and user environment contained within an ACM represents a further advantage of the present invention.

The primary power supply component <b>1830</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment provides electrical energy for the sustained, normal operation of the PCON <b>1800</b> and any ACM coupled to connector <b>1950</b>. The power supply may be of the switching variety well known in the art that receives electrical energy from an AC source <b>1889</b>, such as a wall outlet. Power supply <b>1830</b> reduces the alternating current input voltage, to a number of distinct outputs of differing voltages and current capacities. The outputs of power supply <b>1830</b> are applied to power bus <b>1831</b>. Power bus <b>1831</b> distributes the power supply outputs to the other circuitry within the PCON <b>1800</b>. Bus <b>1831</b> also connects to certain pins of connector <b>1950</b> to provide the electrical power <b>1914</b> for an ACM conveyed by ACM-to-PCON interconnection <b>1900</b>. The isolation of the usually heavy power supply <b>1830</b> from the core computing power and user environment contained within the ACM represents a further advantage of the present invention.

The interface and support component <b>1840</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises peripheral bridge <b>1846</b>, diskette controller <b>1842</b>, IDE controller <b>1848</b>, and signal conveyance paths <b>1841</b>, <b>1843</b>, <b>1844</b>, <b>1845</b>, <b>1847</b> and <b>1849</b>. Peripheral bridge <b>1846</b> couples PCI peripheral bus <b>1841</b> with peripheral busses of other formats such as ISA peripheral bus <b>1845</b> and others <b>1847</b>. PCI and ISA peripheral busses are industry standards, well known and understood in the art. Other peripheral busses <b>1847</b> may include, for example, a bus compliant with the universal serial bus (USB) industry standard. While other embodiments of a peripheral console <b>1800</b> may include a single peripheral bus that is coupled to an attached ACM via ACM-to-PCON interconnection <b>1900</b>, such as PCI bus <b>1841</b>, this embodiment includes peripheral bridge <b>1846</b> to establish additional busses <b>1845</b>, <b>1847</b>. The additional busses <b>1845</b>, <b>1847</b> permit the use of the many low-cost and readily available components compatible with these bus specifications.

Diskette controller <b>1842</b> interfaces a floppy disk drive <b>1854</b> with the CPU component <b>1710</b> of an attached ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>) so that the CPU may control and use the diskette drive <b>1854</b> hardware to store and retrieve data. Diskette controller <b>1842</b> couples to the CPU via a connection to ISA bus <b>1845</b>. Diskette controller <b>1842</b> connects to the diskette drive <b>1854</b> via one of device cables <b>1843</b>.

Similarly, IDE controller <b>1848</b> interfaces a hard disk drive <b>1852</b> and a CDROM drive <b>1856</b> with the CPU component <b>1710</b> of an attached ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>) so that the CPU may control and use the hard disk drive <b>1852</b> and CDROM <b>1856</b> hardware to store and retrieve data. IDE controller <b>1848</b> couples to the CPU via connection to PCI peripheral bus <b>1841</b>. IDE controller <b>1848</b> connects to each of hard disk drive <b>1852</b> and CD-ROM drive <b>1856</b> via one of device cables <b>1843</b>. Some embodiments of PCON <b>1800</b> may take advantage of VLSI integrated circuits such as an 82371SB (PIIX4) integrated circuit from Intel Corporation. An 82371SB integrated circuit includes circuitry for both the peripheral bridge <b>1846</b> and the IDE controller <b>1848</b> in a single package.

The secondary mass storage component <b>1850</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises diskette drive <b>1854</b>, hard disk drive <b>1852</b>, and CD-ROM drive <b>1856</b>. Secondary mass storage <b>1850</b> generally provides low-cost, non-volatile storage for data files which may include software program files. Data files stored on secondary mass storage <b>1850</b> are not part of a computer user's core computing power and environment. Secondary mass storage <b>1850</b> may be used to store, for example, seldom used software programs, software programs that are used only with companion hardware devices installed in the same peripheral console <b>1800</b>, or archival copies of data files that are maintained in primary mass storage <b>1750</b> of an ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>). Storage capacities for secondary mass storage <b>1850</b> devices may vary from the 1.44 megabytes of the 3.5-inch high density diskette drive <b>1854</b>, to more than 10 gigabytes for a large format (5-inch) hard disk drive <b>1852</b>. Hard disk drive <b>1852</b> employs fixed recording media, while diskette drive <b>1854</b> and CD-ROM drive <b>1856</b> employ removable media. Diskette drive <b>1854</b> and hard disk drive <b>1852</b> support both read and write operations (i.e., data stored on their recording media may be both recalled and modified) while CD-ROM drive <b>1856</b> supports only read operations.

The other devices component <b>1860</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises a video capture card. A video capture card accepts analog television signals, such as those complying with the NTSC standard used for television broadcast in the United States, and digitizes picture frames represented by the analog signal for processing by the computer. Video capture cards at present are considered a specialty, i.e., not ubiquitous, component of personal computer systems. Digitized picture information from video capture card <b>1860</b> is carried via signal conveyance path <b>1844</b> to the peripheral interface controller <b>1940</b> which transforms it to the video port signals <b>1917</b> of the ACM-to-PCON interconnection <b>1900</b> for coupling to the advanced graphics processor <b>1752</b> in an attached ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>).

Video capture card <b>1860</b> is merely representative of the many types of “other” devices that may be installed in a PCON to expand the capabilities of the personal computer. Sound cards and laboratory data acquisition cards are other examples. Video capture card <b>1860</b> is shown installed in one of expansion slots <b>1870</b> for coupling to the interface and control circuitry <b>1840</b> of the PCON. Any of other devices <b>1860</b> could be coupled to the interface and control circuitry <b>1840</b> of the PCON by different means, such as direct installation on the circuit board that includes the interface and control circuitry <b>1840</b>; e.g., a motherboard.

The expansion slots component <b>1870</b> of the PCON functional circuitry <b>1801</b> of the presently described embodiment comprises PCI connectors <b>1871</b> and ISA connectors <b>1872</b>. A circuit card may be inserted into one of the connectors <b>1871</b>, <b>1872</b> in order to be operatively coupled with the CPU <b>1710</b> of an attached ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>). Each of connectors <b>1871</b> electrically connects to PCI bus <b>1841</b>, and may receive and hold a printed circuit card which it electrically couples to PCI bus <b>1841</b>. Each of connectors <b>1872</b> electrically connects to ISA bus <b>1845</b>, and may receive and hold a printed circuit card which it electrically couples to ISA bus <b>1845</b>. The PCI <b>1841</b> and ISA <b>1845</b> busses couple to the CPU <b>1710</b> of an attached ACM (shown in <figref idref="DRAWINGS">FIG. 18</figref>) by circuitry already described.

An embodiment of a detachable computing module in accordance with the present invention, for attachment to a peripheral console for forming a fully operational computer system, comprises, an enclosure, a CPU, a memory coupled to said CPU, and a mass storage coupled to said CPU. The module further comprises interconnection circuitry coupled to said CPU, said interconnection circuitry connectable to a peripheral console. The CPU is uncoupled from any primary input circuitry when said interconnection circuitry is disconnected from a peripheral console.

An alternative embodiment of a detachable computing module in accordance with the present invention, for attachment to a peripheral console for forming a fully operational computer system, comprises an enclosure, a CPU, a memory coupled to said CPU, and a mass storage coupled to said CPU. The module further comprises interconnection circuitry coupled to said CPU, said interconnection circuitry connectable to a peripheral console. The CPU is uncoupled from any primary output circuitry when said interconnection circuitry is disconnected from a peripheral console.

Various modifications to the preferred embodiment can be made without departing from the spirit and scope of the invention. (A limited number of modifications have already been described in the preceding discussion.) For example, a particular embodiment may insert another layer of bus bridging between the CPU bridge and the Peripheral bridge. This may be desirable if, for example, a vendor wants to implement a proprietary, general-purpose bus having intermediate performance characteristics that fall between those of the high-performance general purpose bus originating at the CPU, and the slower general purpose PCI bus. Thus, the foregoing description is not intended to limit the invention as set forth.

In most embodiments, the ACM includes an enclosure such as the one described with the following components, which should not be limiting:

1) A CPU with cache memory;

2) Core logic device or means;

3) Main memory;

4) A single primary Hard Disk Drive (“HDD”) that has a security program;

5) Flash memory with system BIOS and programmable user password;

6) Operating System, application software, data files on primary HDD;

7) An interface device and connectors to peripheral console;

8) A software controllable mechanical lock, lock control means, and other accessories.

The ACM connects to a peripheral console with power supply, a display device, an input device, and other elements. Some details of these elements with the present security system are described in more detail below.

<figref idref="DRAWINGS">FIG. 22</figref> is a simplified layout diagram of a security system for a computer system according to an embodiment of the present invention. This diagram is merely an illustration and should not limit the scope of the claims herein. One of ordinary skill in the an would recognize other variations, modifications. and alternatives. The layout diagram illustrates the top-view of the module <b>22</b>, where the backside components (e.g., Host Interface Controller) are depicted in dashed lines. The layout diagram has a first portion, which includes a central processing unit (“CPU”) module <b>2200</b>, and a second portion, which includes a hard drive module <b>2220</b>. A common printed circuit board <b>2237</b> houses these modules and the like. Among other features, the ACM includes the central processing unit module <b>2200</b> with a cache memory <b>2205</b>, which is coupled to a north bridge unit <b>2221</b>, and a host interface controller <b>2201</b>. The host interface controller includes a lock control <b>2203</b>. As shown, the CPU module is disposed on a first portion of the attached computer module, and couples to connectors <b>2217</b>. Here, the CPU module is spatially located near connector <b>2217</b>.

The CPU module can use a suitable microprocessing unit, microcontroller, digital signal processor, and the like. In a specific embodiment, the CPU module uses, for example, a 400 MHz Pentium II microprocessor module from Intel Corporation and like microprocessors from AMD Corporation, Cyrix Corporation (now National Semiconductor Corporation), and others. In other aspects, the microprocessor can be one such as the Compaq Computer Corporation Alpha Chip, Apple Computer Corporation PowerPC G3 processor, and the like. Further, higher speed processors are contemplated in other embodiments as technology increases in the future.

In the CPU module, host interface controller <b>2201</b> is coupled to BIOS/flash memory <b>2205</b>. Additionally, the host interface controller is coupled to a clock control logic, a configuration signal, and a peripheral bus. The present invention has a host interface controller that has lock control <b>2203</b> to provide security features to the present ACM. Furthermore, the present invention uses a flash memory that includes codes to provide password protection or other electronic security methods.

The second portion of the attached computer module has the hard drive module <b>2220</b>. Among other elements, the hard drive module includes north bridge <b>2221</b>, graphics accelerator <b>2223</b>, graphics memory <b>2225</b>, a power controller <b>2227</b>, an IDE controller <b>2229</b>, and other components. Adjacent to and in parallel alignment with the hard drive module is a personal computer interface (“PCI”) bus <b>2231</b>, <b>2232</b>. A power regulator <b>435</b> is disposed near the PCI bus.

In a specific embodiment, north bridge unit <b>2221</b> often couples to a computer memory, to the graphics accelerator <b>2223</b>, to the IDE controller, and to the host interface controller via the PCI bus. Graphics accelerator <b>2223</b> typically couples to a graphics memory <b>2223</b>, and other elements. IDE controller <b>2229</b> generally supports and provides timing signals necessary for the IDE bus. In the present embodiment, the IDE controller is embodied as a 643U2 PCI-to IDE chip from CMD Technology, for example. Other types of buses than IDE are contemplated, for example EIDE, SCSI, 1394, and the like in alternative embodiments of the present invention.

The hard drive module or mass storage unit <b>2220</b> typically includes a computer operating system, application software program files, data files, and the like. In a specific embodiment, the computer operating system may be the Windows98 operating system from Microsoft Corporation of Redmond Wash. Other operating systems, such as WindowsNT, MacOS8, Unix, and the like are also contemplated in alternative embodiments of the present invention. Further, some typical application software programs can include Office98 by Microsoft Corporation, Corel Perfect Suite by Corel, and others. Hard disk module <b>2220</b> includes a hard disk drive. The hard disk drive, however, can also be replaced by removable hard disk drives, read/write CD ROMs, flash memory, floppy disk drives, and the like. A small form factor, for example 2.5″, is currently contemplated, however, other form factors, such as PC card, and the like are also contemplated. Mass storage unit <b>2240</b> may also support other interfaces than IDE. Among other features, the computer system includes an ACM with security protection. The ACM connects to the console, which has at least the following elements, which should not be limiting.

1) Connection to input devices, e.g. keyboard or mouse;

2) Connection to display devices, e.g. Monitor;

3) Add-on means, e.g. PCI add-on slots;

4) Removable storage media subsystem, e.g. Floppy drive, CDROM drive;

5) Communication device, e.g. LAN or modem;

6) An interface device and connectors to ACM;

7) A computer module bay with a notch in the frame for ACM's lock; and

8) Power supply and other accessories.

As noted, the computer module bay is an opening in a peripheral console that receives the ACM. The computer module bay provides mechanical support and protection to ACM. The module bay also includes, among other elements, a variety of thermal components for heat dissipation, a frame that provides connector alignment, and a lock engagement, which secures the ACM to the console. The bay also has a printed circuit board to mount and mate the connector from the ACM to the console. The connector provides an interface between the ACM and other accessories.

<figref idref="DRAWINGS">FIG. 23</figref> is a simplified block diagram <b>2300</b> of a security system for a computer module according to an embodiment of the present invention. This diagram is merely an illustration and should not limit the scope of the claims herein. One of ordinary skill in the art would recognize other variations, modifications, and alternatives. The block diagram <b>2300</b> has a variety of features such as those noted above, as well as others. In the present diagram, different reference numerals are used to show the operation of the present system.

The block diagram is an attached computer module <b>2300</b>. The module <b>2300</b> has a central processing unit, which communicates to a north bridge <b>2341</b>, by way of a CPU bus <b>2327</b>. The north bridge couples to main memory <b>2323</b> via memory bus <b>2329</b>. The main memory can be any suitable high speed memory device or devices such as dynamic random access memory (“DRAM”) integrated circuits and others. The DRAM includes at least 32 Meg. or 64 Meg. and greater of memory, but can also be less depending upon the application. Alternatively, the main memory can be coupled directly with the CPU in some embodiments. The north bridge also couples to a graphics subsystem <b>2315</b> via bus <b>2342</b>. The graphics subsystem can include a graphics accelerator, graphics memory, and other devices. Graphics subsystem transmits a video signal to an interface connector, which couples to a display, for example.

The attached computer module also includes a primary hard disk drive that serves as a main memory unit for programs and the like. The hard disk can be any suitable drive that has at least 2 GB and greater. As merely an example, the hard disk is a Marathon 2250 (2.25 GB, 2½ inch drive) product made by Seagate Corporation of Scotts Valley, but can be others. The hard disk communicates to the north bridge by way of a hard disk drive controller and bus lines <b>2302</b> and <b>2331</b>. The hard disk drive controller couples to the north bridge by way of the host PCI bus, which connects bus <b>2337</b> to the north bridge. The hard disk includes computer codes that implement a security program according to the present invention. Details of the security program are provided below.

The attached computer module also has a flash memory device <b>2305</b> with a BIOS. The flash memory device <b>2305</b> also has codes for a user password that can be stored in the device. The flash memory device generally pewits the storage of such password without a substantial use of power, even when disconnected. As merely an example, the flash memory device has at least 4 Meg. or greater of memory, or 16 Meg. or greater of memory. A host interface controller <b>2307</b> communications to the north bridge via bus <b>2335</b> and host PCI bus. The host interface controller also has a Lock control <b>2309</b>, which couples to a lock. The lock is attached to the module and has a manual override to the lock on the host interface controller in some embodiments. Host interface controller <b>2307</b> communicates to the console using bus <b>2311</b>, which couples to connection <b>2313</b>.

In a preferred embodiment, the present invention uses a password protection scheme to electronically prevent unauthorized access to the computer module. The present password protection scheme uses a combination of software, which is a portion of the security program, and a user password, which can be stored in the flash memory device <b>505</b>. By way of the flash memory device, the password does not become erased by way of power failure or the lock. The password is substantially fixed in code, which cannot be easily erased. Should the user desire to change the password, it can readily be changed by erasing the code, which is stored in flash memory and a new code (i.e., password) is written into the flash memory. An example of a flash memory device can include a Intel Flash 28F800F3 series flash, which is available in 8 Mbit and 16 Mbit designs. Other types of flash devices can also be used, however. Details of a password protection method are further explained below by way of the FIGS.

In a specific embodiment, the present invention also includes a real-time clock <b>510</b> in the ACM, but is not limited. The real-time clock can be implemented using a reference oscillator 14.31818 MHz <b>508</b> that couples to a real-time clock circuit. The real-time clock circuit can be in the host interface controller. An energy source <b>506</b> such as a battery can be used to keep the real-time clock circuit running even when the ACM has been removed from the console. The real-time clock can be used by a security program to perform a variety of functions. As merely an example, these functions include: (1) fixed time period in which the ACM can be used, e.g., ACM cannot be used at night; (2) programmed ACM to be used after certain date, e.g., high security procedure during owner's vacation or non use period; (3) other uses similar to a programmable time lock. Further details of the present real-time clock are described in the application listed under Ser. No. 09/183,816 noted above.

In still a further embodiment, the present invention also includes a permanent password or user identification code to identify the computer module. In one embodiment, the permanent password or user code is stored in a flash memory device. Alternatively, the permanent password or user code is stored in the central processing unit. The password or user code can be placed in the device upon manufacture of such device. Alternatively, the password or user code can be placed in the device by a one time programming techniques using, for example, fuses or the like. The present password or user code provides a permanent “finger print” on the device, which is generally hardware. The permanent finger print can be used for identification purposes for allowing the user of the hardware to access the hardware itself, as well as other systems. These other systems include local and wide area networks. Alternatively, the systems can also include one or more servers. The present password and user identification can be quite important for electronic commerce applications and the like. In one or more embodiments, the permanent password or user code can be combined with the password on flash memory for the security program, which is described below in more detail.

In one aspect of the invention, the user password is programmable. The password can be programmable by way of the security program. The password can be stored in a flash memory device within the ACM. Accordingly, the user of the ACM and the console would need to have the user password in order to access the ACM. In the present aspect, the combination of a security program and user password can provide the user a wide variety of security functions as follows:

1) Auto-lock capability when ACM is inserted into CMB;

2) Access privilege of program and data;

3) Password matching for ACM removal; and

4) Automatic HDD lock out if tempering is detected.

In still a further embodiment, the present invention also includes a method for reading a permanent password or user identification code to identify the computer module. In one embodiment, the permanent password or user code is stored in a flash memory device. Alternatively, the permanent password or user code is stored in the central processing unit. The password or user code can be placed in the device upon manufacture of such device. Alternatively, the password or user code can be placed in the device by a one time programming techniques using, for example, fuses or the like. The present password or user code provides a permanent “finger print” on the device, which is generally hardware. The permanent finger print can be used for identification purposes for allowing the user of the hardware to access the hardware itself, as well as other systems. These other systems include local and wide area networks. Alternatively, the systems can also include one or more servers. The present method allows a third party confirm the user by way of the permanent password or user code. The present password and user identification can be quite important for electronic commerce applications and the like, which verify the user code or password. In one or more embodiments, the permanent password or user code can be combined with the password on flash memory for the security program.

The above embodiments are described generally in terms of hardware and software. It will be recognized, however, that the functionality of the hardware can be further combined or even separated. The functionality of the software can also be further combined or even separated. Hardware can be replaced, at times, with software. Software can be replaced, at times, with hardware. Accordingly, the present embodiments should not be construed as limiting the scope of the claims here. One of ordinary skill in the art would recognize other variations, modifications, and alternatives.

While the above is a full description of the specific embodiments, various modifications, alternative constructions and equivalents may be used. Therefore, the above description and illustrations should not be taken as limiting the scope of the present invention which is defined by the appended claims.